Import modules¶
In [1]:
from google.colab import drive
drive.mount('/content/drive')
Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount("/content/drive", force_remount=True).
In [2]:
!pip install tensorflow --upgrade
Requirement already up-to-date: tensorflow in /usr/local/lib/python3.6/dist-packages (2.0.0) Requirement already satisfied, skipping upgrade: astor>=0.6.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow) (0.8.1) Requirement already satisfied, skipping upgrade: termcolor>=1.1.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow) (1.1.0) Requirement already satisfied, skipping upgrade: keras-preprocessing>=1.0.5 in /usr/local/lib/python3.6/dist-packages (from tensorflow) (1.1.0) Requirement already satisfied, skipping upgrade: opt-einsum>=2.3.2 in /usr/local/lib/python3.6/dist-packages (from tensorflow) (3.1.0) Requirement already satisfied, skipping upgrade: wheel>=0.26 in /usr/local/lib/python3.6/dist-packages (from tensorflow) (0.33.6) Requirement already satisfied, skipping upgrade: google-pasta>=0.1.6 in /usr/local/lib/python3.6/dist-packages (from tensorflow) (0.1.8) Requirement already satisfied, skipping upgrade: keras-applications>=1.0.8 in /usr/local/lib/python3.6/dist-packages (from tensorflow) (1.0.8) Requirement already satisfied, skipping upgrade: six>=1.10.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow) (1.12.0) Requirement already satisfied, skipping upgrade: absl-py>=0.7.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow) (0.8.1) Requirement already satisfied, skipping upgrade: grpcio>=1.8.6 in /usr/local/lib/python3.6/dist-packages (from tensorflow) (1.15.0) Requirement already satisfied, skipping upgrade: tensorboard<2.1.0,>=2.0.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow) (2.0.2) Requirement already satisfied, skipping upgrade: tensorflow-estimator<2.1.0,>=2.0.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow) (2.0.1) Requirement already satisfied, skipping upgrade: protobuf>=3.6.1 in /usr/local/lib/python3.6/dist-packages (from tensorflow) (3.10.0) Requirement already satisfied, skipping upgrade: wrapt>=1.11.1 in /usr/local/lib/python3.6/dist-packages (from tensorflow) (1.11.2) Requirement already satisfied, skipping upgrade: numpy<2.0,>=1.16.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow) (1.17.4) Requirement already satisfied, skipping upgrade: gast==0.2.2 in /usr/local/lib/python3.6/dist-packages (from tensorflow) (0.2.2) Requirement already satisfied, skipping upgrade: h5py in /usr/local/lib/python3.6/dist-packages (from keras-applications>=1.0.8->tensorflow) (2.8.0) Requirement already satisfied, skipping upgrade: markdown>=2.6.8 in /usr/local/lib/python3.6/dist-packages (from tensorboard<2.1.0,>=2.0.0->tensorflow) (3.1.1) Requirement already satisfied, skipping upgrade: werkzeug>=0.11.15 in /usr/local/lib/python3.6/dist-packages (from tensorboard<2.1.0,>=2.0.0->tensorflow) (0.16.0) Requirement already satisfied, skipping upgrade: google-auth<2,>=1.6.3 in /usr/local/lib/python3.6/dist-packages (from tensorboard<2.1.0,>=2.0.0->tensorflow) (1.8.2) Requirement already satisfied, skipping upgrade: google-auth-oauthlib<0.5,>=0.4.1 in /usr/local/lib/python3.6/dist-packages (from tensorboard<2.1.0,>=2.0.0->tensorflow) (0.4.1) Requirement already satisfied, skipping upgrade: requests<3,>=2.21.0 in /usr/local/lib/python3.6/dist-packages (from tensorboard<2.1.0,>=2.0.0->tensorflow) (2.21.0) Requirement already satisfied, skipping upgrade: setuptools>=41.0.0 in /usr/local/lib/python3.6/dist-packages (from tensorboard<2.1.0,>=2.0.0->tensorflow) (42.0.2) Requirement already satisfied, skipping upgrade: cachetools<3.2,>=2.0.0 in /usr/local/lib/python3.6/dist-packages (from google-auth<2,>=1.6.3->tensorboard<2.1.0,>=2.0.0->tensorflow) (3.1.1) Requirement already satisfied, skipping upgrade: rsa<4.1,>=3.1.4 in /usr/local/lib/python3.6/dist-packages (from google-auth<2,>=1.6.3->tensorboard<2.1.0,>=2.0.0->tensorflow) (4.0) Requirement already satisfied, skipping upgrade: pyasn1-modules>=0.2.1 in /usr/local/lib/python3.6/dist-packages (from google-auth<2,>=1.6.3->tensorboard<2.1.0,>=2.0.0->tensorflow) (0.2.7) Requirement already satisfied, skipping upgrade: requests-oauthlib>=0.7.0 in /usr/local/lib/python3.6/dist-packages (from google-auth-oauthlib<0.5,>=0.4.1->tensorboard<2.1.0,>=2.0.0->tensorflow) (1.3.0) Requirement already satisfied, skipping upgrade: idna<2.9,>=2.5 in /usr/local/lib/python3.6/dist-packages (from requests<3,>=2.21.0->tensorboard<2.1.0,>=2.0.0->tensorflow) (2.8) Requirement already satisfied, skipping upgrade: urllib3<1.25,>=1.21.1 in /usr/local/lib/python3.6/dist-packages (from requests<3,>=2.21.0->tensorboard<2.1.0,>=2.0.0->tensorflow) (1.24.3) Requirement already satisfied, skipping upgrade: chardet<3.1.0,>=3.0.2 in /usr/local/lib/python3.6/dist-packages (from requests<3,>=2.21.0->tensorboard<2.1.0,>=2.0.0->tensorflow) (3.0.4) Requirement already satisfied, skipping upgrade: certifi>=2017.4.17 in /usr/local/lib/python3.6/dist-packages (from requests<3,>=2.21.0->tensorboard<2.1.0,>=2.0.0->tensorflow) (2019.11.28) Requirement already satisfied, skipping upgrade: pyasn1>=0.1.3 in /usr/local/lib/python3.6/dist-packages (from rsa<4.1,>=3.1.4->google-auth<2,>=1.6.3->tensorboard<2.1.0,>=2.0.0->tensorflow) (0.4.8) Requirement already satisfied, skipping upgrade: oauthlib>=3.0.0 in /usr/local/lib/python3.6/dist-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib<0.5,>=0.4.1->tensorboard<2.1.0,>=2.0.0->tensorflow) (3.1.0)
In [0]:
import tensorflow as tf
# Set different random seeds for comparison
tf.random.set_seed(0)
#tf.random.set_seed(2019)
from tensorflow.python.keras.preprocessing import image as kp_image
from tensorflow.python.keras import models
from tensorflow.python.keras import losses
from tensorflow.python.keras import layers
from tensorflow.python.keras import backend as K
In [0]:
import IPython.display
import matplotlib.pyplot as plt
import matplotlib as mpl
mpl.rcParams['figure.figsize'] = (12,12)
mpl.rcParams['axes.grid'] = False
import numpy as np
from PIL import Image
import time
import functools
Input Content and Style Images¶
In [0]:
# Set up image path
# Use different types of content and style images
content_path1 = '/content/drive/My Drive/AML Project/people.jpg'
content_path2 = '/content/drive/My Drive/AML Project/puppy-dog.jpg'
content_path3 = '/content/drive/My Drive/AML Project/Shanghai.jpg'
content_path4 = '/content/drive/My Drive/AML Project/dancing.jpg'
style_path1 = '/content/drive/My Drive/AML Project/picasso.jpg'
style_path2 = '/content/drive/My Drive/AML Project/great_wave.jpg'
style_path3 = '/content/drive/My Drive/AML Project/Van_Gogh.jpg'
style_path4 = '/content/drive/My Drive/AML Project/cezanne.jpg'
In [0]:
def load_img(path_to_img):
max_dim = 512
img = Image.open(path_to_img)
img_size = max(img.size)
scale = max_dim/img_size
img = img.resize((round(img.size[0]*scale), round(img.size[1]*scale)), Image.ANTIALIAS)
img = kp_image.img_to_array(img)
img = np.expand_dims(img, axis=0)
return img
In [0]:
def imshow(img, title=None):
# Remove the batch dimension
out = np.squeeze(img, axis=0)
out = out.astype('uint8')
plt.imshow(out)
if title is not None:
plt.title(title)
plt.imshow(out)
In [8]:
# Show the input content and style images
plt.figure(figsize=(12,12))
content_image1 = load_img(content_path1).astype('uint8')
style_image1 = load_img(style_path1).astype('uint8')
content_image2 = load_img(content_path2).astype('uint8')
style_image2 = load_img(style_path2).astype('uint8')
content_image3 = load_img(content_path3).astype('uint8')
style_image3 = load_img(style_path3).astype('uint8')
content_image4 = load_img(content_path4).astype('uint8')
style_image4 = load_img(style_path4).astype('uint8')
plt.subplot(4, 2, 1)
imshow(content_image1, 'Content Image1')
plt.subplot(4, 2, 2)
imshow(style_image1, 'Style Image1')
plt.subplot(4, 2, 3)
imshow(content_image2, 'Content Image2')
plt.subplot(4, 2, 4)
imshow(style_image2, 'Style Image2')
plt.subplot(4, 2, 5)
imshow(content_image3, 'Content Image3')
plt.subplot(4, 2, 6)
imshow(style_image3, 'Style Image3')
plt.subplot(4, 2, 7)
imshow(content_image4, 'Content Image4')
plt.subplot(4, 2, 8)
imshow(style_image4, 'Style Image4')
plt.show()
Preprocess data¶
In [0]:
def load_and_process_img(path_to_img):
img = load_img(path_to_img)
# Preprocess raw images to make it suitable to be used by pre-trained VGG19 model
output = tf.keras.applications.vgg19.preprocess_input(img)
return output
In [0]:
def deprocess_img(processed_img):
x = processed_img.copy()
if len(x.shape) == 4:
x = np.squeeze(x, 0)
assert len(x.shape) == 3, ("Input to deprocess image must be an image of "
"dimension [1, height, width, channel] or [height, width, channel]")
if len(x.shape) != 3:
raise ValueError("Invalid input to deprocessing image")
# Perform the inverse of the preprocessiing step
# VGG networks are trained on image with each channel normalized by mean = [103.939, 116.779, 123.68] and with channels BGR
x[:, :, 0] += 103.939
x[:, :, 1] += 116.779
x[:, :, 2] += 123.68
x = x[:, :, ::-1]
x = np.clip(x, 0, 255).astype('uint8')
return x
Define content and style representations¶
In [11]:
# Use weights of pre-trained VGG19 to build our neural network
vgg = tf.keras.applications.VGG19(include_top=False, weights="/content/drive/My Drive/AML Project/vgg19_weights_tf_dim_ordering_tf_kernels_notop.h5")
for layer in vgg.layers:
print(layer.name)
input_1 block1_conv1 block1_conv2 block1_pool block2_conv1 block2_conv2 block2_pool block3_conv1 block3_conv2 block3_conv3 block3_conv4 block3_pool block4_conv1 block4_conv2 block4_conv3 block4_conv4 block4_pool block5_conv1 block5_conv2 block5_conv3 block5_conv4 block5_pool
In [0]:
# List of Content and Style layers to be considered for calculation of Content and Style Loss
# Content layer where will pull our feature maps
# Try different Content layers to see the difference
# In the reference paper, the author stated that "We therefore refer to the feature responses in higher layers of the network as the content representation."
content_layers1 = ['block5_conv2']
content_layers2 = ['block5_conv3']
content_layers3 = ['block4_conv2']
content_layers4 = ['block3_conv2']
# Style layer of interest
style_layers = ['block1_conv1',
'block2_conv1',
'block3_conv1',
'block4_conv1',
'block5_conv1']
num_content_layers1 = len(content_layers1)
num_content_layers2 = len(content_layers2)
num_content_layers3 = len(content_layers3)
num_content_layers4 = len(content_layers4)
num_style_layers = len(style_layers)
Build Model¶
In [0]:
def get_model(content_layers = content_layers1):
""" Creates our model with access to intermediate layers."""
# Load our model and use pre-trained VGG19 weights
vgg = tf.keras.applications.vgg19.VGG19(include_top=False, weights="/content/drive/My Drive/AML Project/vgg19_weights_tf_dim_ordering_tf_kernels_notop.h5")
vgg.trainable = False
# Get output layers corresponding to style and content layers
style_outputs = [vgg.get_layer(name).output for name in style_layers]
content_outputs = [vgg.get_layer(name).output for name in content_layers]
model_outputs = style_outputs + content_outputs
# Build model
model = tf.keras.Model([vgg.input], model_outputs)
return model
Define and create loss functions (both style and content loss)¶
Style Loss¶
In [0]:
def gram_matrix(input_tensor):
# Since the input tensor is a 3D array of size height * width * channels, we reshape it to a 2D array of (height * width) * channels
channels = int(input_tensor.shape[-1])
# -1 means the size of height * width
a = tf.reshape(input_tensor, [-1, channels])
b = tf.reshape(input_tensor, [-1, channels])
# Gram matrix G_ij is the inner product, and G_ij has dimension N_l * N_l
# Thus, gram matrix is the inner product of the transpose matrix of A and A, which will have size N_l * N_l
gram = tf.matmul(a, b, transpose_a=True)
return gram
In [0]:
def get_style_loss(base_style, gram_target):
# Get height, width, and num filters of each layer
# We scale the loss at a given layer by the size of the feature map and the number of filters
height, width, channels = base_style.get_shape().as_list()
gram_style = gram_matrix(base_style)
# gram_target will be processed as a gram_matrix in the run_style_transfer step
# Style Loss = 1/(4 * (N_l * M_l)^2) * sum_i,j((G_ij - A_ij)^2), where M_l is the height times the width of the feature map
return tf.reduce_sum((gram_style - gram_target)**2)/(4*(channels * height * width)**2)
Content Loss¶
In [0]:
def get_content_loss(content, target):
# Content Loss = 1/2 * sum_i,j((F_ij - P_ij)^2)
return tf.reduce_sum((content - target)**2)/2
Total Loss¶
In [0]:
def get_feature_representations(model, content_path, style_path):
# Load our images
content_image = load_and_process_img(content_path)
style_image = load_and_process_img(style_path)
style_outputs = model(style_image)
content_outputs = model(content_image)
# Get the style and content feature representations from our model
style_features = [style_layer[0] for style_layer in style_outputs[:num_style_layers]]
content_features = [content_layer[0] for content_layer in content_outputs[num_style_layers:]]
return style_features, content_features
In [0]:
def compute_loss(model, num_content_layers, loss_weights, init_image, gram_style_features, content_features):
style_weight, content_weight = loss_weights
generated = model(init_image)
style_output_features = generated[:num_style_layers]
content_output_features = generated[num_style_layers:]
style_score = 0
content_score = 0
# Accumulate style losses from all layers
# We use equal weights for each content and style loss layer
weight_per_style_layer = 1.0 / float(num_style_layers)
for target_style, comb_style in zip(gram_style_features, style_output_features):
style_score += weight_per_style_layer * get_style_loss(comb_style[0], target_style)
# Accumulate content losses from all layers
weight_per_content_layer = 1.0 / float(num_content_layers)
for target_content, comb_content in zip(content_features, content_output_features):
content_score += weight_per_content_layer * get_content_loss(comb_content[0], target_content)
# Get total loss
loss = style_weight * style_score + content_weight * content_score
return loss, style_score, content_score
In [0]:
def compute_grads(cfg):
# Use tf.GradientTape() to compute the gradient
with tf.GradientTape() as tape:
all_loss = compute_loss(**cfg)
# Compute gradients
total_loss = all_loss[0]
return tape.gradient(total_loss, cfg['init_image']), all_loss
Style Transfer¶
In [0]:
def run_style_transfer(content_path,
style_path,
content_layers=content_layers1,
num_iterations=1000,
content_weight=1,
style_weight=1e3):
# Part of the function was inspired by the reference article published by TensorFlow
display_num = 1
#display_num = 10
num_content_layers = len(content_layers)
model = get_model(content_layers)
for layer in model.layers:
layer.trainable = False
# Get the style and content feature representations
style_features, content_features = get_feature_representations(model, content_path, style_path)
gram_style_features = [gram_matrix(style_feature) for style_feature in style_features]
# Set initial content image
init_image = load_and_process_img(content_path)
init_image = tf.Variable(init_image, dtype=tf.float32)
# Create our optimizer and Adam will work here (The paper suggested L-BFGS but we saw many other referrence still used Adam and Adam also worked)
opt = tf.keras.optimizers.Adam(learning_rate=9, beta_1=0.99, epsilon=1e-1)
# For displaying intermediate images
iter_count = 1
# Store our best result
best_loss, best_img = float('inf'), None
# Create lists to store all three losses
content_loss = []
style_loss = []
total_loss = []
# Create a nice config
loss_weights = (style_weight, content_weight)
cfg = {
'model': model,
'num_content_layers': num_content_layers,
'loss_weights': loss_weights,
'init_image': init_image,
'gram_style_features': gram_style_features,
'content_features': content_features
}
# For displaying
plt.figure(figsize=(12, 12))
num_rows = (num_iterations / display_num) // 5
start_time = time.time()
global_start = time.time()
norm_means = np.array([103.939, 116.779, 123.68])
min_vals = -norm_means
max_vals = 255 - norm_means
# Create a list to store all pictures we want in iteration process
output = []
iter_list = [100,200,300,400,500,600,700,800,900,1000]
for i in range(num_iterations):
grads, all_loss = compute_grads(cfg)
loss, style_score, content_score = all_loss
# grads, _ = tf.clip_by_global_norm(grads, 5.0)
opt.apply_gradients([(grads, init_image)])
clipped = tf.clip_by_value(init_image, min_vals, max_vals)
init_image.assign(clipped)
end_time = time.time()
if loss < best_loss:
# Update best loss and best image from total loss
best_loss = loss
best_img = init_image.numpy()
if i % display_num == 0:
print('Iteration: {}'.format(i))
print('Total loss: {:.4e}, '
'style loss: {:.4e}, '
'content loss: {:.4e}, '
'time: {:.4f}s'.format(loss, style_score, content_score, time.time() - start_time))
start_time = time.time()
content_loss.append(content_score)
style_loss.append(style_score)
total_loss.append(loss)
if (i+1) in iter_list:
output.append(best_img)
print('Total time: {:.4f}s'.format(time.time() - global_start))
return best_img, total_loss, style_loss, content_loss, output
Visualization¶
In [0]:
def show_results(best_img, content_path, style_path, show_large_final=True):
plt.figure(figsize=(10, 5))
content = load_img(content_path)
style = load_img(style_path)
plt.subplot(1, 2, 1)
imshow(content, 'Content Image')
plt.subplot(1, 2, 2)
imshow(style, 'Style Image')
if show_large_final:
plt.figure(figsize=(10, 10))
plt.imshow(deprocess_img(best_img))
plt.title('Output Image')
plt.show()
In [22]:
# Seed 0
import time
start = time.time()
best, best_loss, style_score, content_score, output = run_style_transfer(content_path3, style_path3, content_layers=content_layers4, num_iterations=1000, content_weight=1, style_weight = 1e3)
show_results(best, content_path3, style_path3)
end = time.time()
print("Total time: {:.1f}".format(end-start))
Iteration: 0 Total loss: 6.2955e+12, style loss: 6.2955e+09, content loss: 0.0000e+00, time: 3.8978s Iteration: 1 Total loss: 2.1630e+12, style loss: 2.1324e+09, content loss: 3.0632e+10, time: 3.9728s Iteration: 2 Total loss: 2.4872e+12, style loss: 2.4260e+09, content loss: 6.1202e+10, time: 3.9835s Iteration: 3 Total loss: 1.8947e+12, style loss: 1.8253e+09, content loss: 6.9429e+10, time: 3.9834s Iteration: 4 Total loss: 1.6219e+12, style loss: 1.5467e+09, content loss: 7.5269e+10, time: 4.0218s Iteration: 5 Total loss: 1.2543e+12, style loss: 1.1728e+09, content loss: 8.1492e+10, time: 4.0237s Iteration: 6 Total loss: 1.1217e+12, style loss: 1.0339e+09, content loss: 8.7705e+10, time: 3.9792s Iteration: 7 Total loss: 1.0310e+12, style loss: 9.3871e+08, content loss: 9.2315e+10, time: 3.9621s Iteration: 8 Total loss: 9.4406e+11, style loss: 8.4831e+08, content loss: 9.5752e+10, time: 3.9513s Iteration: 9 Total loss: 9.1374e+11, style loss: 8.1458e+08, content loss: 9.9159e+10, time: 3.9844s Iteration: 10 Total loss: 8.6276e+11, style loss: 7.5964e+08, content loss: 1.0312e+11, time: 3.9608s Iteration: 11 Total loss: 8.0270e+11, style loss: 6.9513e+08, content loss: 1.0757e+11, time: 3.9766s Iteration: 12 Total loss: 7.9375e+11, style loss: 6.8182e+08, content loss: 1.1193e+11, time: 3.9669s Iteration: 13 Total loss: 8.0174e+11, style loss: 6.8637e+08, content loss: 1.1538e+11, time: 3.9559s Iteration: 14 Total loss: 7.7441e+11, style loss: 6.5680e+08, content loss: 1.1761e+11, time: 3.9448s Iteration: 15 Total loss: 7.4883e+11, style loss: 6.2970e+08, content loss: 1.1913e+11, time: 3.9975s Iteration: 16 Total loss: 7.4385e+11, style loss: 6.2325e+08, content loss: 1.2060e+11, time: 3.9867s Iteration: 17 Total loss: 7.2604e+11, style loss: 6.0364e+08, content loss: 1.2240e+11, time: 3.9561s Iteration: 18 Total loss: 6.9078e+11, style loss: 5.6622e+08, content loss: 1.2456e+11, time: 3.9435s Iteration: 19 Total loss: 6.6681e+11, style loss: 5.3995e+08, content loss: 1.2686e+11, time: 3.9786s Iteration: 20 Total loss: 6.6371e+11, style loss: 5.3477e+08, content loss: 1.2893e+11, time: 3.9890s Iteration: 21 Total loss: 6.5827e+11, style loss: 5.2784e+08, content loss: 1.3043e+11, time: 4.0018s Iteration: 22 Total loss: 6.3770e+11, style loss: 5.0640e+08, content loss: 1.3130e+11, time: 3.9759s Iteration: 23 Total loss: 6.1908e+11, style loss: 4.8727e+08, content loss: 1.3181e+11, time: 3.9428s Iteration: 24 Total loss: 6.1551e+11, style loss: 4.8321e+08, content loss: 1.3230e+11, time: 4.0320s Iteration: 25 Total loss: 6.1625e+11, style loss: 4.8323e+08, content loss: 1.3302e+11, time: 3.9840s Iteration: 26 Total loss: 6.0628e+11, style loss: 4.7222e+08, content loss: 1.3407e+11, time: 3.9877s Iteration: 27 Total loss: 5.8800e+11, style loss: 4.5266e+08, content loss: 1.3534e+11, time: 3.9788s Iteration: 28 Total loss: 5.7433e+11, style loss: 4.3767e+08, content loss: 1.3667e+11, time: 3.9904s Iteration: 29 Total loss: 5.6879e+11, style loss: 4.3099e+08, content loss: 1.3779e+11, time: 3.9386s Iteration: 30 Total loss: 5.6287e+11, style loss: 4.2437e+08, content loss: 1.3851e+11, time: 3.9722s Iteration: 31 Total loss: 5.5075e+11, style loss: 4.1197e+08, content loss: 1.3878e+11, time: 3.9611s Iteration: 32 Total loss: 5.3746e+11, style loss: 3.9872e+08, content loss: 1.3874e+11, time: 3.9684s Iteration: 33 Total loss: 5.2997e+11, style loss: 3.9136e+08, content loss: 1.3860e+11, time: 4.0342s Iteration: 34 Total loss: 5.2715e+11, style loss: 3.8856e+08, content loss: 1.3858e+11, time: 3.9901s Iteration: 35 Total loss: 5.2301e+11, style loss: 3.8422e+08, content loss: 1.3879e+11, time: 3.9815s Iteration: 36 Total loss: 5.1536e+11, style loss: 3.7615e+08, content loss: 1.3921e+11, time: 3.9864s Iteration: 37 Total loss: 5.0676e+11, style loss: 3.6701e+08, content loss: 1.3975e+11, time: 3.9818s Iteration: 38 Total loss: 4.9932e+11, style loss: 3.5907e+08, content loss: 1.4026e+11, time: 3.9561s Iteration: 39 Total loss: 4.9175e+11, style loss: 3.5115e+08, content loss: 1.4060e+11, time: 4.0111s Iteration: 40 Total loss: 4.8270e+11, style loss: 3.4201e+08, content loss: 1.4069e+11, time: 3.9538s Iteration: 41 Total loss: 4.7373e+11, style loss: 3.3315e+08, content loss: 1.4058e+11, time: 3.9726s Iteration: 42 Total loss: 4.6719e+11, style loss: 3.2684e+08, content loss: 1.4035e+11, time: 3.9821s Iteration: 43 Total loss: 4.6289e+11, style loss: 3.2277e+08, content loss: 1.4012e+11, time: 4.0038s Iteration: 44 Total loss: 4.5840e+11, style loss: 3.1841e+08, content loss: 1.3999e+11, time: 4.0051s Iteration: 45 Total loss: 4.5229e+11, style loss: 3.1228e+08, content loss: 1.4001e+11, time: 4.0046s Iteration: 46 Total loss: 4.4604e+11, style loss: 3.0588e+08, content loss: 1.4016e+11, time: 4.0275s Iteration: 47 Total loss: 4.4152e+11, style loss: 3.0117e+08, content loss: 1.4036e+11, time: 3.9832s Iteration: 48 Total loss: 4.3801e+11, style loss: 2.9751e+08, content loss: 1.4051e+11, time: 3.9771s Iteration: 49 Total loss: 4.3263e+11, style loss: 2.9212e+08, content loss: 1.4051e+11, time: 3.9864s Iteration: 50 Total loss: 4.2424e+11, style loss: 2.8390e+08, content loss: 1.4034e+11, time: 3.9963s Iteration: 51 Total loss: 4.1530e+11, style loss: 2.7528e+08, content loss: 1.4002e+11, time: 4.0045s Iteration: 52 Total loss: 4.0880e+11, style loss: 2.6917e+08, content loss: 1.3964e+11, time: 3.9844s Iteration: 53 Total loss: 4.0471e+11, style loss: 2.6544e+08, content loss: 1.3927e+11, time: 3.9613s Iteration: 54 Total loss: 4.0058e+11, style loss: 2.6156e+08, content loss: 1.3902e+11, time: 3.9915s Iteration: 55 Total loss: 3.9488e+11, style loss: 2.5598e+08, content loss: 1.3889e+11, time: 3.9501s Iteration: 56 Total loss: 3.8892e+11, style loss: 2.5004e+08, content loss: 1.3888e+11, time: 3.9828s Iteration: 57 Total loss: 3.8501e+11, style loss: 2.4611e+08, content loss: 1.3891e+11, time: 3.9553s Iteration: 58 Total loss: 3.8309e+11, style loss: 2.4420e+08, content loss: 1.3889e+11, time: 3.9540s Iteration: 59 Total loss: 3.8070e+11, style loss: 2.4194e+08, content loss: 1.3876e+11, time: 3.9580s Iteration: 60 Total loss: 3.7584e+11, style loss: 2.3736e+08, content loss: 1.3847e+11, time: 3.9717s Iteration: 61 Total loss: 3.6946e+11, style loss: 2.3139e+08, content loss: 1.3807e+11, time: 3.9746s Iteration: 62 Total loss: 3.6396e+11, style loss: 2.2636e+08, content loss: 1.3761e+11, time: 3.9726s Iteration: 63 Total loss: 3.6023e+11, style loss: 2.2306e+08, content loss: 1.3717e+11, time: 3.9732s Iteration: 64 Total loss: 3.5688e+11, style loss: 2.2006e+08, content loss: 1.3683e+11, time: 4.0088s Iteration: 65 Total loss: 3.5233e+11, style loss: 2.1572e+08, content loss: 1.3661e+11, time: 4.0150s Iteration: 66 Total loss: 3.4681e+11, style loss: 2.1032e+08, content loss: 1.3649e+11, time: 4.0374s Iteration: 67 Total loss: 3.4199e+11, style loss: 2.0556e+08, content loss: 1.3644e+11, time: 4.0195s Iteration: 68 Total loss: 3.3879e+11, style loss: 2.0241e+08, content loss: 1.3638e+11, time: 3.9977s Iteration: 69 Total loss: 3.3640e+11, style loss: 2.0016e+08, content loss: 1.3624e+11, time: 4.0176s Iteration: 70 Total loss: 3.3333e+11, style loss: 1.9733e+08, content loss: 1.3600e+11, time: 3.9386s Iteration: 71 Total loss: 3.2921e+11, style loss: 1.9355e+08, content loss: 1.3565e+11, time: 3.9414s Iteration: 72 Total loss: 3.2503e+11, style loss: 1.8978e+08, content loss: 1.3525e+11, time: 3.9881s Iteration: 73 Total loss: 3.2169e+11, style loss: 1.8684e+08, content loss: 1.3485e+11, time: 3.9837s Iteration: 74 Total loss: 3.1893e+11, style loss: 1.8442e+08, content loss: 1.3451e+11, time: 3.9679s Iteration: 75 Total loss: 3.1594e+11, style loss: 1.8168e+08, content loss: 1.3425e+11, time: 3.9566s Iteration: 76 Total loss: 3.1247e+11, style loss: 1.7838e+08, content loss: 1.3408e+11, time: 3.9591s Iteration: 77 Total loss: 3.0899e+11, style loss: 1.7503e+08, content loss: 1.3396e+11, time: 3.9772s Iteration: 78 Total loss: 3.0584e+11, style loss: 1.7201e+08, content loss: 1.3384e+11, time: 3.9435s Iteration: 79 Total loss: 3.0286e+11, style loss: 1.6921e+08, content loss: 1.3365e+11, time: 3.9762s Iteration: 80 Total loss: 2.9980e+11, style loss: 1.6642e+08, content loss: 1.3339e+11, time: 3.9869s Iteration: 81 Total loss: 2.9675e+11, style loss: 1.6372e+08, content loss: 1.3303e+11, time: 3.9853s Iteration: 82 Total loss: 2.9406e+11, style loss: 1.6143e+08, content loss: 1.3263e+11, time: 3.9967s Iteration: 83 Total loss: 2.9154e+11, style loss: 1.5931e+08, content loss: 1.3223e+11, time: 3.9745s Iteration: 84 Total loss: 2.8862e+11, style loss: 1.5675e+08, content loss: 1.3187e+11, time: 4.0169s Iteration: 85 Total loss: 2.8509e+11, style loss: 1.5350e+08, content loss: 1.3159e+11, time: 3.9980s Iteration: 86 Total loss: 2.8157e+11, style loss: 1.5018e+08, content loss: 1.3139e+11, time: 4.0143s Iteration: 87 Total loss: 2.7882e+11, style loss: 1.4760e+08, content loss: 1.3122e+11, time: 3.9688s Iteration: 88 Total loss: 2.7680e+11, style loss: 1.4575e+08, content loss: 1.3105e+11, time: 3.9595s Iteration: 89 Total loss: 2.7466e+11, style loss: 1.4383e+08, content loss: 1.3084e+11, time: 3.9573s Iteration: 90 Total loss: 2.7185e+11, style loss: 1.4131e+08, content loss: 1.3055e+11, time: 3.9590s Iteration: 91 Total loss: 2.6886e+11, style loss: 1.3867e+08, content loss: 1.3020e+11, time: 3.9847s Iteration: 92 Total loss: 2.6657e+11, style loss: 1.3676e+08, content loss: 1.2981e+11, time: 3.9868s Iteration: 93 Total loss: 2.6502e+11, style loss: 1.3558e+08, content loss: 1.2943e+11, time: 3.9298s Iteration: 94 Total loss: 2.6323e+11, style loss: 1.3412e+08, content loss: 1.2911e+11, time: 3.9306s Iteration: 95 Total loss: 2.6041e+11, style loss: 1.3157e+08, content loss: 1.2884e+11, time: 3.9894s Iteration: 96 Total loss: 2.5690e+11, style loss: 1.2828e+08, content loss: 1.2863e+11, time: 3.9427s Iteration: 97 Total loss: 2.5385e+11, style loss: 1.2540e+08, content loss: 1.2845e+11, time: 3.9505s Iteration: 98 Total loss: 2.5177e+11, style loss: 1.2352e+08, content loss: 1.2826e+11, time: 3.9645s Iteration: 99 Total loss: 2.5011e+11, style loss: 1.2209e+08, content loss: 1.2802e+11, time: 3.9606s Iteration: 100 Total loss: 2.4804e+11, style loss: 1.2033e+08, content loss: 1.2771e+11, time: 3.9640s Iteration: 101 Total loss: 2.4557e+11, style loss: 1.1822e+08, content loss: 1.2734e+11, time: 3.9327s Iteration: 102 Total loss: 2.4341e+11, style loss: 1.1645e+08, content loss: 1.2696e+11, time: 3.9766s Iteration: 103 Total loss: 2.4190e+11, style loss: 1.1531e+08, content loss: 1.2659e+11, time: 3.9636s Iteration: 104 Total loss: 2.4059e+11, style loss: 1.1431e+08, content loss: 1.2628e+11, time: 3.9448s Iteration: 105 Total loss: 2.3878e+11, style loss: 1.1274e+08, content loss: 1.2604e+11, time: 3.9823s Iteration: 106 Total loss: 2.3646e+11, style loss: 1.1060e+08, content loss: 1.2586e+11, time: 4.0051s Iteration: 107 Total loss: 2.3416e+11, style loss: 1.0846e+08, content loss: 1.2570e+11, time: 3.9850s Iteration: 108 Total loss: 2.3223e+11, style loss: 1.0670e+08, content loss: 1.2553e+11, time: 3.9303s Iteration: 109 Total loss: 2.3049e+11, style loss: 1.0518e+08, content loss: 1.2531e+11, time: 4.0057s Iteration: 110 Total loss: 2.2867e+11, style loss: 1.0364e+08, content loss: 1.2503e+11, time: 4.0282s Iteration: 111 Total loss: 2.2681e+11, style loss: 1.0212e+08, content loss: 1.2470e+11, time: 3.9493s Iteration: 112 Total loss: 2.2520e+11, style loss: 1.0085e+08, content loss: 1.2435e+11, time: 3.9548s Iteration: 113 Total loss: 2.2382e+11, style loss: 9.9803e+07, content loss: 1.2402e+11, time: 3.9580s Iteration: 114 Total loss: 2.2238e+11, style loss: 9.8649e+07, content loss: 1.2373e+11, time: 3.9485s Iteration: 115 Total loss: 2.2076e+11, style loss: 9.7263e+07, content loss: 1.2349e+11, time: 3.9995s Iteration: 116 Total loss: 2.1910e+11, style loss: 9.5805e+07, content loss: 1.2329e+11, time: 3.9353s Iteration: 117 Total loss: 2.1758e+11, style loss: 9.4484e+07, content loss: 1.2309e+11, time: 3.9701s Iteration: 118 Total loss: 2.1609e+11, style loss: 9.3220e+07, content loss: 1.2287e+11, time: 3.9893s Iteration: 119 Total loss: 2.1441e+11, style loss: 9.1824e+07, content loss: 1.2259e+11, time: 3.9487s Iteration: 120 Total loss: 2.1262e+11, style loss: 9.0365e+07, content loss: 1.2226e+11, time: 3.9462s Iteration: 121 Total loss: 2.1107e+11, style loss: 8.9181e+07, content loss: 1.2189e+11, time: 3.9674s Iteration: 122 Total loss: 2.0989e+11, style loss: 8.8350e+07, content loss: 1.2154e+11, time: 4.0015s Iteration: 123 Total loss: 2.0871e+11, style loss: 8.7489e+07, content loss: 1.2122e+11, time: 3.9574s Iteration: 124 Total loss: 2.0722e+11, style loss: 8.6254e+07, content loss: 1.2097e+11, time: 3.9851s Iteration: 125 Total loss: 2.0553e+11, style loss: 8.4766e+07, content loss: 1.2076e+11, time: 4.0424s Iteration: 126 Total loss: 2.0411e+11, style loss: 8.3514e+07, content loss: 1.2060e+11, time: 3.9761s Iteration: 127 Total loss: 2.0311e+11, style loss: 8.2679e+07, content loss: 1.2043e+11, time: 4.0007s Iteration: 128 Total loss: 2.0210e+11, style loss: 8.1883e+07, content loss: 1.2021e+11, time: 3.9566s Iteration: 129 Total loss: 2.0066e+11, style loss: 8.0719e+07, content loss: 1.1994e+11, time: 3.9475s Iteration: 130 Total loss: 1.9894e+11, style loss: 7.9322e+07, content loss: 1.1961e+11, time: 3.9874s Iteration: 131 Total loss: 1.9745e+11, style loss: 7.8195e+07, content loss: 1.1926e+11, time: 3.9539s Iteration: 132 Total loss: 1.9644e+11, style loss: 7.7531e+07, content loss: 1.1890e+11, time: 3.9632s Iteration: 133 Total loss: 1.9552e+11, style loss: 7.6934e+07, content loss: 1.1859e+11, time: 3.9833s Iteration: 134 Total loss: 1.9430e+11, style loss: 7.5976e+07, content loss: 1.1833e+11, time: 3.9612s Iteration: 135 Total loss: 1.9286e+11, style loss: 7.4739e+07, content loss: 1.1812e+11, time: 3.9348s Iteration: 136 Total loss: 1.9160e+11, style loss: 7.3664e+07, content loss: 1.1793e+11, time: 3.9896s Iteration: 137 Total loss: 1.9070e+11, style loss: 7.2953e+07, content loss: 1.1775e+11, time: 3.9755s Iteration: 138 Total loss: 1.8988e+11, style loss: 7.2343e+07, content loss: 1.1754e+11, time: 3.9481s Iteration: 139 Total loss: 1.8879e+11, style loss: 7.1512e+07, content loss: 1.1728e+11, time: 3.9817s Iteration: 140 Total loss: 1.8749e+11, style loss: 7.0499e+07, content loss: 1.1699e+11, time: 3.9552s Iteration: 141 Total loss: 1.8626e+11, style loss: 6.9575e+07, content loss: 1.1668e+11, time: 3.9547s Iteration: 142 Total loss: 1.8521e+11, style loss: 6.8806e+07, content loss: 1.1640e+11, time: 3.9955s Iteration: 143 Total loss: 1.8415e+11, style loss: 6.8002e+07, content loss: 1.1615e+11, time: 3.9817s Iteration: 144 Total loss: 1.8301e+11, style loss: 6.7066e+07, content loss: 1.1594e+11, time: 3.9533s Iteration: 145 Total loss: 1.8189e+11, style loss: 6.6125e+07, content loss: 1.1576e+11, time: 4.0620s Iteration: 146 Total loss: 1.8090e+11, style loss: 6.5321e+07, content loss: 1.1558e+11, time: 4.0063s Iteration: 147 Total loss: 1.8000e+11, style loss: 6.4632e+07, content loss: 1.1537e+11, time: 3.9893s Iteration: 148 Total loss: 1.7909e+11, style loss: 6.3967e+07, content loss: 1.1512e+11, time: 4.0121s Iteration: 149 Total loss: 1.7819e+11, style loss: 6.3360e+07, content loss: 1.1483e+11, time: 3.9521s Iteration: 150 Total loss: 1.7740e+11, style loss: 6.2872e+07, content loss: 1.1453e+11, time: 3.9359s Iteration: 151 Total loss: 1.7663e+11, style loss: 6.2398e+07, content loss: 1.1423e+11, time: 3.9801s Iteration: 152 Total loss: 1.7569e+11, style loss: 6.1732e+07, content loss: 1.1396e+11, time: 3.9652s Iteration: 153 Total loss: 1.7459e+11, style loss: 6.0857e+07, content loss: 1.1374e+11, time: 3.9395s Iteration: 154 Total loss: 1.7356e+11, style loss: 6.0017e+07, content loss: 1.1354e+11, time: 3.9796s Iteration: 155 Total loss: 1.7271e+11, style loss: 5.9363e+07, content loss: 1.1335e+11, time: 3.9277s Iteration: 156 Total loss: 1.7191e+11, style loss: 5.8766e+07, content loss: 1.1314e+11, time: 3.9598s Iteration: 157 Total loss: 1.7098e+11, style loss: 5.8070e+07, content loss: 1.1291e+11, time: 3.9603s Iteration: 158 Total loss: 1.7003e+11, style loss: 5.7396e+07, content loss: 1.1263e+11, time: 3.9327s Iteration: 159 Total loss: 1.6931e+11, style loss: 5.6967e+07, content loss: 1.1235e+11, time: 3.9410s Iteration: 160 Total loss: 1.6878e+11, style loss: 5.6715e+07, content loss: 1.1206e+11, time: 3.9722s Iteration: 161 Total loss: 1.6812e+11, style loss: 5.6309e+07, content loss: 1.1181e+11, time: 3.9576s Iteration: 162 Total loss: 1.6717e+11, style loss: 5.5580e+07, content loss: 1.1159e+11, time: 3.9515s Iteration: 163 Total loss: 1.6614e+11, style loss: 5.4747e+07, content loss: 1.1140e+11, time: 3.9930s Iteration: 164 Total loss: 1.6534e+11, style loss: 5.4132e+07, content loss: 1.1121e+11, time: 3.9725s Iteration: 165 Total loss: 1.6474e+11, style loss: 5.3728e+07, content loss: 1.1101e+11, time: 3.9424s Iteration: 166 Total loss: 1.6404e+11, style loss: 5.3257e+07, content loss: 1.1078e+11, time: 4.0172s Iteration: 167 Total loss: 1.6315e+11, style loss: 5.2640e+07, content loss: 1.1051e+11, time: 4.0234s Iteration: 168 Total loss: 1.6231e+11, style loss: 5.2088e+07, content loss: 1.1022e+11, time: 3.9304s Iteration: 169 Total loss: 1.6168e+11, style loss: 5.1743e+07, content loss: 1.0994e+11, time: 3.9579s Iteration: 170 Total loss: 1.6112e+11, style loss: 5.1442e+07, content loss: 1.0968e+11, time: 3.9228s Iteration: 171 Total loss: 1.6039e+11, style loss: 5.0948e+07, content loss: 1.0944e+11, time: 3.9431s Iteration: 172 Total loss: 1.5952e+11, style loss: 5.0284e+07, content loss: 1.0923e+11, time: 3.9777s Iteration: 173 Total loss: 1.5870e+11, style loss: 4.9675e+07, content loss: 1.0903e+11, time: 3.9341s Iteration: 174 Total loss: 1.5804e+11, style loss: 4.9226e+07, content loss: 1.0881e+11, time: 3.9157s Iteration: 175 Total loss: 1.5740e+11, style loss: 4.8842e+07, content loss: 1.0856e+11, time: 4.0048s Iteration: 176 Total loss: 1.5671e+11, style loss: 4.8434e+07, content loss: 1.0828e+11, time: 3.9227s Iteration: 177 Total loss: 1.5602e+11, style loss: 4.8041e+07, content loss: 1.0798e+11, time: 3.9433s Iteration: 178 Total loss: 1.5538e+11, style loss: 4.7692e+07, content loss: 1.0769e+11, time: 3.9579s Iteration: 179 Total loss: 1.5474e+11, style loss: 4.7310e+07, content loss: 1.0743e+11, time: 3.9461s Iteration: 180 Total loss: 1.5405e+11, style loss: 4.6842e+07, content loss: 1.0721e+11, time: 3.9369s Iteration: 181 Total loss: 1.5340e+11, style loss: 4.6381e+07, content loss: 1.0701e+11, time: 3.9592s Iteration: 182 Total loss: 1.5284e+11, style loss: 4.6011e+07, content loss: 1.0683e+11, time: 3.9525s Iteration: 183 Total loss: 1.5228e+11, style loss: 4.5657e+07, content loss: 1.0662e+11, time: 3.9442s Iteration: 184 Total loss: 1.5161e+11, style loss: 4.5218e+07, content loss: 1.0639e+11, time: 3.9556s Iteration: 185 Total loss: 1.5089e+11, style loss: 4.4773e+07, content loss: 1.0612e+11, time: 3.9419s Iteration: 186 Total loss: 1.5032e+11, style loss: 4.4478e+07, content loss: 1.0584e+11, time: 4.0599s Iteration: 187 Total loss: 1.4986e+11, style loss: 4.4297e+07, content loss: 1.0556e+11, time: 4.1067s Iteration: 188 Total loss: 1.4932e+11, style loss: 4.4011e+07, content loss: 1.0531e+11, time: 4.0009s Iteration: 189 Total loss: 1.4862e+11, style loss: 4.3535e+07, content loss: 1.0508e+11, time: 3.8838s Iteration: 190 Total loss: 1.4793e+11, style loss: 4.3053e+07, content loss: 1.0488e+11, time: 4.0000s Iteration: 191 Total loss: 1.4741e+11, style loss: 4.2735e+07, content loss: 1.0467e+11, time: 3.9368s Iteration: 192 Total loss: 1.4694e+11, style loss: 4.2493e+07, content loss: 1.0445e+11, time: 3.9167s Iteration: 193 Total loss: 1.4636e+11, style loss: 4.2160e+07, content loss: 1.0420e+11, time: 3.9464s Iteration: 194 Total loss: 1.4568e+11, style loss: 4.1766e+07, content loss: 1.0392e+11, time: 3.9355s Iteration: 195 Total loss: 1.4510e+11, style loss: 4.1476e+07, content loss: 1.0362e+11, time: 3.8113s Iteration: 196 Total loss: 1.4463e+11, style loss: 4.1293e+07, content loss: 1.0334e+11, time: 3.8304s Iteration: 197 Total loss: 1.4414e+11, style loss: 4.1054e+07, content loss: 1.0309e+11, time: 3.8257s Iteration: 198 Total loss: 1.4355e+11, style loss: 4.0683e+07, content loss: 1.0286e+11, time: 3.8057s Iteration: 199 Total loss: 1.4295e+11, style loss: 4.0297e+07, content loss: 1.0266e+11, time: 3.8117s Iteration: 200 Total loss: 1.4246e+11, style loss: 4.0007e+07, content loss: 1.0245e+11, time: 3.8266s Iteration: 201 Total loss: 1.4198e+11, style loss: 3.9761e+07, content loss: 1.0222e+11, time: 3.8284s Iteration: 202 Total loss: 1.4144e+11, style loss: 3.9474e+07, content loss: 1.0197e+11, time: 3.8372s Iteration: 203 Total loss: 1.4087e+11, style loss: 3.9179e+07, content loss: 1.0169e+11, time: 3.8325s Iteration: 204 Total loss: 1.4034e+11, style loss: 3.8936e+07, content loss: 1.0141e+11, time: 3.8165s Iteration: 205 Total loss: 1.3984e+11, style loss: 3.8702e+07, content loss: 1.0114e+11, time: 3.8255s Iteration: 206 Total loss: 1.3930e+11, style loss: 3.8395e+07, content loss: 1.0090e+11, time: 3.8935s Iteration: 207 Total loss: 1.3874e+11, style loss: 3.8049e+07, content loss: 1.0069e+11, time: 3.8891s Iteration: 208 Total loss: 1.3823e+11, style loss: 3.7749e+07, content loss: 1.0048e+11, time: 3.9059s Iteration: 209 Total loss: 1.3776e+11, style loss: 3.7492e+07, content loss: 1.0027e+11, time: 3.8560s Iteration: 210 Total loss: 1.3725e+11, style loss: 3.7222e+07, content loss: 1.0003e+11, time: 3.8385s Iteration: 211 Total loss: 1.3673e+11, style loss: 3.6960e+07, content loss: 9.9774e+10, time: 3.7946s Iteration: 212 Total loss: 1.3628e+11, style loss: 3.6777e+07, content loss: 9.9505e+10, time: 3.8349s Iteration: 213 Total loss: 1.3587e+11, style loss: 3.6625e+07, content loss: 9.9243e+10, time: 3.8198s Iteration: 214 Total loss: 1.3538e+11, style loss: 3.6374e+07, content loss: 9.9003e+10, time: 3.8363s Iteration: 215 Total loss: 1.3480e+11, style loss: 3.6011e+07, content loss: 9.8791e+10, time: 3.8118s Iteration: 216 Total loss: 1.3427e+11, style loss: 3.5676e+07, content loss: 9.8596e+10, time: 3.8199s Iteration: 217 Total loss: 1.3385e+11, style loss: 3.5449e+07, content loss: 9.8404e+10, time: 3.8329s Iteration: 218 Total loss: 1.3344e+11, style loss: 3.5238e+07, content loss: 9.8197e+10, time: 3.8226s Iteration: 219 Total loss: 1.3293e+11, style loss: 3.4964e+07, content loss: 9.7967e+10, time: 3.8491s Iteration: 220 Total loss: 1.3243e+11, style loss: 3.4710e+07, content loss: 9.7719e+10, time: 3.8344s Iteration: 221 Total loss: 1.3204e+11, style loss: 3.4570e+07, content loss: 9.7467e+10, time: 3.8477s Iteration: 222 Total loss: 1.3169e+11, style loss: 3.4463e+07, content loss: 9.7228e+10, time: 3.8334s Iteration: 223 Total loss: 1.3126e+11, style loss: 3.4246e+07, content loss: 9.7010e+10, time: 3.8225s Iteration: 224 Total loss: 1.3074e+11, style loss: 3.3932e+07, content loss: 9.6809e+10, time: 3.9492s Iteration: 225 Total loss: 1.3026e+11, style loss: 3.3651e+07, content loss: 9.6612e+10, time: 3.9210s Iteration: 226 Total loss: 1.2985e+11, style loss: 3.3445e+07, content loss: 9.6402e+10, time: 3.8849s Iteration: 227 Total loss: 1.2942e+11, style loss: 3.3252e+07, content loss: 9.6170e+10, time: 3.8986s Iteration: 228 Total loss: 1.2897e+11, style loss: 3.3053e+07, content loss: 9.5920e+10, time: 3.8600s Iteration: 229 Total loss: 1.2855e+11, style loss: 3.2890e+07, content loss: 9.5665e+10, time: 3.9038s Iteration: 230 Total loss: 1.2817e+11, style loss: 3.2750e+07, content loss: 9.5421e+10, time: 3.8333s Iteration: 231 Total loss: 1.2777e+11, style loss: 3.2566e+07, content loss: 9.5199e+10, time: 3.8240s Iteration: 232 Total loss: 1.2733e+11, style loss: 3.2331e+07, content loss: 9.4998e+10, time: 3.8650s Iteration: 233 Total loss: 1.2692e+11, style loss: 3.2111e+07, content loss: 9.4805e+10, time: 3.8501s Iteration: 234 Total loss: 1.2653e+11, style loss: 3.1928e+07, content loss: 9.4605e+10, time: 3.8398s Iteration: 235 Total loss: 1.2612e+11, style loss: 3.1734e+07, content loss: 9.4385e+10, time: 3.8276s Iteration: 236 Total loss: 1.2568e+11, style loss: 3.1535e+07, content loss: 9.4145e+10, time: 3.8337s Iteration: 237 Total loss: 1.2529e+11, style loss: 3.1392e+07, content loss: 9.3895e+10, time: 3.8182s Iteration: 238 Total loss: 1.2494e+11, style loss: 3.1290e+07, content loss: 9.3653e+10, time: 3.8542s Iteration: 239 Total loss: 1.2457e+11, style loss: 3.1131e+07, content loss: 9.3434e+10, time: 3.8393s Iteration: 240 Total loss: 1.2413e+11, style loss: 3.0892e+07, content loss: 9.3239e+10, time: 3.8399s Iteration: 241 Total loss: 1.2373e+11, style loss: 3.0672e+07, content loss: 9.3060e+10, time: 3.8459s Iteration: 242 Total loss: 1.2340e+11, style loss: 3.0520e+07, content loss: 9.2880e+10, time: 3.8134s Iteration: 243 Total loss: 1.2305e+11, style loss: 3.0370e+07, content loss: 9.2684e+10, time: 3.8345s Iteration: 244 Total loss: 1.2265e+11, style loss: 3.0181e+07, content loss: 9.2466e+10, time: 3.8386s Iteration: 245 Total loss: 1.2225e+11, style loss: 3.0013e+07, content loss: 9.2233e+10, time: 3.8535s Iteration: 246 Total loss: 1.2191e+11, style loss: 2.9910e+07, content loss: 9.1999e+10, time: 3.8274s Iteration: 247 Total loss: 1.2158e+11, style loss: 2.9799e+07, content loss: 9.1781e+10, time: 3.9274s Iteration: 248 Total loss: 1.2119e+11, style loss: 2.9606e+07, content loss: 9.1585e+10, time: 4.0679s Iteration: 249 Total loss: 1.2079e+11, style loss: 2.9380e+07, content loss: 9.1406e+10, time: 3.9809s Iteration: 250 Total loss: 1.2044e+11, style loss: 2.9208e+07, content loss: 9.1231e+10, time: 3.9607s Iteration: 251 Total loss: 1.2013e+11, style loss: 2.9082e+07, content loss: 9.1045e+10, time: 3.9123s Iteration: 252 Total loss: 1.1978e+11, style loss: 2.8943e+07, content loss: 9.0840e+10, time: 3.8980s Iteration: 253 Total loss: 1.1941e+11, style loss: 2.8793e+07, content loss: 9.0620e+10, time: 4.3041s Iteration: 254 Total loss: 1.1907e+11, style loss: 2.8669e+07, content loss: 9.0397e+10, time: 4.0010s Iteration: 255 Total loss: 1.1874e+11, style loss: 2.8552e+07, content loss: 9.0185e+10, time: 4.0574s Iteration: 256 Total loss: 1.1839e+11, style loss: 2.8398e+07, content loss: 8.9993e+10, time: 3.9240s Iteration: 257 Total loss: 1.1804e+11, style loss: 2.8224e+07, content loss: 8.9816e+10, time: 3.9350s Iteration: 258 Total loss: 1.1771e+11, style loss: 2.8070e+07, content loss: 8.9643e+10, time: 3.9968s Iteration: 259 Total loss: 1.1738e+11, style loss: 2.7925e+07, content loss: 8.9459e+10, time: 4.0651s Iteration: 260 Total loss: 1.1703e+11, style loss: 2.7776e+07, content loss: 8.9258e+10, time: 4.0208s Iteration: 261 Total loss: 1.1670e+11, style loss: 2.7659e+07, content loss: 8.9043e+10, time: 4.0350s Iteration: 262 Total loss: 1.1641e+11, style loss: 2.7584e+07, content loss: 8.8827e+10, time: 3.9955s Iteration: 263 Total loss: 1.1611e+11, style loss: 2.7487e+07, content loss: 8.8623e+10, time: 4.0186s Iteration: 264 Total loss: 1.1576e+11, style loss: 2.7323e+07, content loss: 8.8438e+10, time: 3.9093s Iteration: 265 Total loss: 1.1542e+11, style loss: 2.7147e+07, content loss: 8.8269e+10, time: 3.8531s Iteration: 266 Total loss: 1.1513e+11, style loss: 2.7020e+07, content loss: 8.8106e+10, time: 3.8988s Iteration: 267 Total loss: 1.1485e+11, style loss: 2.6911e+07, content loss: 8.7935e+10, time: 3.8742s Iteration: 268 Total loss: 1.1452e+11, style loss: 2.6769e+07, content loss: 8.7750e+10, time: 3.9009s Iteration: 269 Total loss: 1.1418e+11, style loss: 2.6628e+07, content loss: 8.7552e+10, time: 3.8696s Iteration: 270 Total loss: 1.1389e+11, style loss: 2.6539e+07, content loss: 8.7354e+10, time: 3.9118s Iteration: 271 Total loss: 1.1363e+11, style loss: 2.6462e+07, content loss: 8.7164e+10, time: 3.8617s Iteration: 272 Total loss: 1.1332e+11, style loss: 2.6328e+07, content loss: 8.6990e+10, time: 3.8421s Iteration: 273 Total loss: 1.1299e+11, style loss: 2.6159e+07, content loss: 8.6827e+10, time: 3.8820s Iteration: 274 Total loss: 1.1269e+11, style loss: 2.6024e+07, content loss: 8.6665e+10, time: 3.8309s Iteration: 275 Total loss: 1.1241e+11, style loss: 2.5921e+07, content loss: 8.6493e+10, time: 3.8980s Iteration: 276 Total loss: 1.1212e+11, style loss: 2.5809e+07, content loss: 8.6306e+10, time: 3.8499s Iteration: 277 Total loss: 1.1181e+11, style loss: 2.5700e+07, content loss: 8.6111e+10, time: 3.8491s Iteration: 278 Total loss: 1.1154e+11, style loss: 2.5619e+07, content loss: 8.5918e+10, time: 3.8446s Iteration: 279 Total loss: 1.1127e+11, style loss: 2.5532e+07, content loss: 8.5740e+10, time: 3.8920s Iteration: 280 Total loss: 1.1098e+11, style loss: 2.5403e+07, content loss: 8.5580e+10, time: 3.8498s Iteration: 281 Total loss: 1.1068e+11, style loss: 2.5254e+07, content loss: 8.5430e+10, time: 3.8717s Iteration: 282 Total loss: 1.1040e+11, style loss: 2.5126e+07, content loss: 8.5279e+10, time: 3.8674s Iteration: 283 Total loss: 1.1012e+11, style loss: 2.5008e+07, content loss: 8.5113e+10, time: 3.8872s Iteration: 284 Total loss: 1.0982e+11, style loss: 2.4895e+07, content loss: 8.4929e+10, time: 3.9165s Iteration: 285 Total loss: 1.0955e+11, style loss: 2.4814e+07, content loss: 8.4735e+10, time: 3.8388s Iteration: 286 Total loss: 1.0930e+11, style loss: 2.4754e+07, content loss: 8.4543e+10, time: 3.8744s Iteration: 287 Total loss: 1.0903e+11, style loss: 2.4664e+07, content loss: 8.4366e+10, time: 4.1176s Iteration: 288 Total loss: 1.0874e+11, style loss: 2.4536e+07, content loss: 8.4206e+10, time: 3.9109s Iteration: 289 Total loss: 1.0848e+11, style loss: 2.4421e+07, content loss: 8.4056e+10, time: 3.9960s Iteration: 290 Total loss: 1.0823e+11, style loss: 2.4328e+07, content loss: 8.3904e+10, time: 4.0084s Iteration: 291 Total loss: 1.0796e+11, style loss: 2.4219e+07, content loss: 8.3738e+10, time: 3.8855s Iteration: 292 Total loss: 1.0766e+11, style loss: 2.4100e+07, content loss: 8.3559e+10, time: 3.8599s Iteration: 293 Total loss: 1.0740e+11, style loss: 2.4022e+07, content loss: 8.3374e+10, time: 4.0385s Iteration: 294 Total loss: 1.0717e+11, style loss: 2.3972e+07, content loss: 8.3196e+10, time: 3.9953s Iteration: 295 Total loss: 1.0692e+11, style loss: 2.3887e+07, content loss: 8.3033e+10, time: 3.9380s Iteration: 296 Total loss: 1.0664e+11, style loss: 2.3757e+07, content loss: 8.2884e+10, time: 3.8913s Iteration: 297 Total loss: 1.0638e+11, style loss: 2.3641e+07, content loss: 8.2740e+10, time: 3.8066s Iteration: 298 Total loss: 1.0615e+11, style loss: 2.3561e+07, content loss: 8.2589e+10, time: 3.8656s Iteration: 299 Total loss: 1.0591e+11, style loss: 2.3483e+07, content loss: 8.2424e+10, time: 3.8412s Iteration: 300 Total loss: 1.0564e+11, style loss: 2.3398e+07, content loss: 8.2246e+10, time: 3.8672s Iteration: 301 Total loss: 1.0539e+11, style loss: 2.3329e+07, content loss: 8.2066e+10, time: 3.8035s Iteration: 302 Total loss: 1.0516e+11, style loss: 2.3261e+07, content loss: 8.1895e+10, time: 3.8626s Iteration: 303 Total loss: 1.0490e+11, style loss: 2.3165e+07, content loss: 8.1740e+10, time: 3.8812s Iteration: 304 Total loss: 1.0465e+11, style loss: 2.3053e+07, content loss: 8.1597e+10, time: 3.9138s Iteration: 305 Total loss: 1.0441e+11, style loss: 2.2955e+07, content loss: 8.1456e+10, time: 3.8706s Iteration: 306 Total loss: 1.0417e+11, style loss: 2.2868e+07, content loss: 8.1306e+10, time: 3.8416s Iteration: 307 Total loss: 1.0393e+11, style loss: 2.2788e+07, content loss: 8.1141e+10, time: 3.8899s Iteration: 308 Total loss: 1.0369e+11, style loss: 2.2729e+07, content loss: 8.0966e+10, time: 3.8498s Iteration: 309 Total loss: 1.0347e+11, style loss: 2.2684e+07, content loss: 8.0790e+10, time: 3.8871s Iteration: 310 Total loss: 1.0324e+11, style loss: 2.2614e+07, content loss: 8.0625e+10, time: 3.9283s Iteration: 311 Total loss: 1.0299e+11, style loss: 2.2516e+07, content loss: 8.0474e+10, time: 4.0041s Iteration: 312 Total loss: 1.0276e+11, style loss: 2.2431e+07, content loss: 8.0331e+10, time: 3.9163s Iteration: 313 Total loss: 1.0255e+11, style loss: 2.2365e+07, content loss: 8.0186e+10, time: 3.8631s Iteration: 314 Total loss: 1.0232e+11, style loss: 2.2289e+07, content loss: 8.0032e+10, time: 3.8621s Iteration: 315 Total loss: 1.0208e+11, style loss: 2.2210e+07, content loss: 7.9870e+10, time: 3.8670s Iteration: 316 Total loss: 1.0186e+11, style loss: 2.2160e+07, content loss: 7.9705e+10, time: 3.8854s Iteration: 317 Total loss: 1.0166e+11, style loss: 2.2116e+07, content loss: 7.9547e+10, time: 3.8060s Iteration: 318 Total loss: 1.0144e+11, style loss: 2.2035e+07, content loss: 7.9400e+10, time: 3.8242s Iteration: 319 Total loss: 1.0120e+11, style loss: 2.1935e+07, content loss: 7.9262e+10, time: 3.9543s Iteration: 320 Total loss: 1.0098e+11, style loss: 2.1859e+07, content loss: 7.9123e+10, time: 3.9622s Iteration: 321 Total loss: 1.0078e+11, style loss: 2.1806e+07, content loss: 7.8974e+10, time: 3.9001s Iteration: 322 Total loss: 1.0056e+11, style loss: 2.1747e+07, content loss: 7.8813e+10, time: 3.8474s Iteration: 323 Total loss: 1.0033e+11, style loss: 2.1690e+07, content loss: 7.8644e+10, time: 3.8634s Iteration: 324 Total loss: 1.0013e+11, style loss: 2.1647e+07, content loss: 7.8479e+10, time: 3.9460s Iteration: 325 Total loss: 9.9922e+10, style loss: 2.1595e+07, content loss: 7.8327e+10, time: 3.8692s Iteration: 326 Total loss: 9.9707e+10, style loss: 2.1519e+07, content loss: 7.8187e+10, time: 3.8432s Iteration: 327 Total loss: 9.9492e+10, style loss: 2.1437e+07, content loss: 7.8055e+10, time: 3.8728s Iteration: 328 Total loss: 9.9284e+10, style loss: 2.1365e+07, content loss: 7.7919e+10, time: 3.8340s Iteration: 329 Total loss: 9.9075e+10, style loss: 2.1301e+07, content loss: 7.7773e+10, time: 3.8581s Iteration: 330 Total loss: 9.8869e+10, style loss: 2.1251e+07, content loss: 7.7618e+10, time: 3.8585s Iteration: 331 Total loss: 9.8674e+10, style loss: 2.1211e+07, content loss: 7.7463e+10, time: 3.9200s Iteration: 332 Total loss: 9.8474e+10, style loss: 2.1158e+07, content loss: 7.7316e+10, time: 3.9077s Iteration: 333 Total loss: 9.8262e+10, style loss: 2.1079e+07, content loss: 7.7183e+10, time: 3.8999s Iteration: 334 Total loss: 9.8061e+10, style loss: 2.1003e+07, content loss: 7.7058e+10, time: 3.8325s Iteration: 335 Total loss: 9.7880e+10, style loss: 2.0947e+07, content loss: 7.6934e+10, time: 3.8326s Iteration: 336 Total loss: 9.7688e+10, style loss: 2.0886e+07, content loss: 7.6802e+10, time: 3.8875s Iteration: 337 Total loss: 9.7479e+10, style loss: 2.0818e+07, content loss: 7.6661e+10, time: 3.8287s Iteration: 338 Total loss: 9.7282e+10, style loss: 2.0765e+07, content loss: 7.6516e+10, time: 3.8459s Iteration: 339 Total loss: 9.7102e+10, style loss: 2.0725e+07, content loss: 7.6377e+10, time: 3.8602s Iteration: 340 Total loss: 9.6911e+10, style loss: 2.0663e+07, content loss: 7.6248e+10, time: 3.9066s Iteration: 341 Total loss: 9.6709e+10, style loss: 2.0580e+07, content loss: 7.6129e+10, time: 3.8302s Iteration: 342 Total loss: 9.6521e+10, style loss: 2.0509e+07, content loss: 7.6011e+10, time: 3.8704s Iteration: 343 Total loss: 9.6342e+10, style loss: 2.0456e+07, content loss: 7.5886e+10, time: 3.8727s Iteration: 344 Total loss: 9.6151e+10, style loss: 2.0403e+07, content loss: 7.5748e+10, time: 3.9271s Iteration: 345 Total loss: 9.5954e+10, style loss: 2.0353e+07, content loss: 7.5602e+10, time: 3.8891s Iteration: 346 Total loss: 9.5770e+10, style loss: 2.0314e+07, content loss: 7.5456e+10, time: 3.8913s Iteration: 347 Total loss: 9.5590e+10, style loss: 2.0271e+07, content loss: 7.5319e+10, time: 3.9024s Iteration: 348 Total loss: 9.5404e+10, style loss: 2.0210e+07, content loss: 7.5194e+10, time: 3.9980s Iteration: 349 Total loss: 9.5223e+10, style loss: 2.0147e+07, content loss: 7.5076e+10, time: 3.9147s Iteration: 350 Total loss: 9.5047e+10, style loss: 2.0091e+07, content loss: 7.4956e+10, time: 3.9013s Iteration: 351 Total loss: 9.4868e+10, style loss: 2.0039e+07, content loss: 7.4829e+10, time: 3.9237s Iteration: 352 Total loss: 9.4689e+10, style loss: 1.9995e+07, content loss: 7.4693e+10, time: 3.9962s Iteration: 353 Total loss: 9.4515e+10, style loss: 1.9960e+07, content loss: 7.4555e+10, time: 3.8972s Iteration: 354 Total loss: 9.4340e+10, style loss: 1.9916e+07, content loss: 7.4423e+10, time: 3.9497s Iteration: 355 Total loss: 9.4158e+10, style loss: 1.9858e+07, content loss: 7.4300e+10, time: 3.9281s Iteration: 356 Total loss: 9.3984e+10, style loss: 1.9803e+07, content loss: 7.4181e+10, time: 3.9264s Iteration: 357 Total loss: 9.3819e+10, style loss: 1.9757e+07, content loss: 7.4062e+10, time: 3.9122s Iteration: 358 Total loss: 9.3644e+10, style loss: 1.9709e+07, content loss: 7.3935e+10, time: 3.9232s Iteration: 359 Total loss: 9.3465e+10, style loss: 1.9663e+07, content loss: 7.3801e+10, time: 3.9101s Iteration: 360 Total loss: 9.3301e+10, style loss: 1.9634e+07, content loss: 7.3667e+10, time: 3.9739s Iteration: 361 Total loss: 9.3140e+10, style loss: 1.9601e+07, content loss: 7.3540e+10, time: 3.9219s Iteration: 362 Total loss: 9.2964e+10, style loss: 1.9540e+07, content loss: 7.3424e+10, time: 3.9189s Iteration: 363 Total loss: 9.2785e+10, style loss: 1.9468e+07, content loss: 7.3317e+10, time: 3.9154s Iteration: 364 Total loss: 9.2624e+10, style loss: 1.9414e+07, content loss: 7.3210e+10, time: 3.8825s Iteration: 365 Total loss: 9.2465e+10, style loss: 1.9369e+07, content loss: 7.3096e+10, time: 3.8587s Iteration: 366 Total loss: 9.2295e+10, style loss: 1.9322e+07, content loss: 7.2974e+10, time: 3.8976s Iteration: 367 Total loss: 9.2130e+10, style loss: 1.9283e+07, content loss: 7.2847e+10, time: 3.9208s Iteration: 368 Total loss: 9.1976e+10, style loss: 1.9251e+07, content loss: 7.2725e+10, time: 3.8632s Iteration: 369 Total loss: 9.1817e+10, style loss: 1.9206e+07, content loss: 7.2611e+10, time: 3.9028s Iteration: 370 Total loss: 9.1655e+10, style loss: 1.9151e+07, content loss: 7.2504e+10, time: 3.8784s Iteration: 371 Total loss: 9.1498e+10, style loss: 1.9101e+07, content loss: 7.2397e+10, time: 3.9095s Iteration: 372 Total loss: 9.1341e+10, style loss: 1.9057e+07, content loss: 7.2283e+10, time: 3.9957s Iteration: 373 Total loss: 9.1181e+10, style loss: 1.9020e+07, content loss: 7.2162e+10, time: 3.9620s Iteration: 374 Total loss: 9.1030e+10, style loss: 1.8992e+07, content loss: 7.2038e+10, time: 3.9096s Iteration: 375 Total loss: 9.0882e+10, style loss: 1.8963e+07, content loss: 7.1919e+10, time: 3.9335s Iteration: 376 Total loss: 9.0727e+10, style loss: 1.8916e+07, content loss: 7.1811e+10, time: 3.8776s Iteration: 377 Total loss: 9.0570e+10, style loss: 1.8860e+07, content loss: 7.1711e+10, time: 3.8905s Iteration: 378 Total loss: 9.0424e+10, style loss: 1.8812e+07, content loss: 7.1611e+10, time: 3.8956s Iteration: 379 Total loss: 9.0276e+10, style loss: 1.8770e+07, content loss: 7.1506e+10, time: 3.9153s Iteration: 380 Total loss: 9.0122e+10, style loss: 1.8729e+07, content loss: 7.1393e+10, time: 3.9171s Iteration: 381 Total loss: 8.9974e+10, style loss: 1.8699e+07, content loss: 7.1275e+10, time: 3.9303s Iteration: 382 Total loss: 8.9833e+10, style loss: 1.8672e+07, content loss: 7.1161e+10, time: 3.9323s Iteration: 383 Total loss: 8.9683e+10, style loss: 1.8627e+07, content loss: 7.1055e+10, time: 3.9588s Iteration: 384 Total loss: 8.9528e+10, style loss: 1.8570e+07, content loss: 7.0958e+10, time: 4.0311s Iteration: 385 Total loss: 8.9384e+10, style loss: 1.8522e+07, content loss: 7.0862e+10, time: 3.9521s Iteration: 386 Total loss: 8.9244e+10, style loss: 1.8483e+07, content loss: 7.0761e+10, time: 3.9091s Iteration: 387 Total loss: 8.9097e+10, style loss: 1.8444e+07, content loss: 7.0652e+10, time: 3.8740s Iteration: 388 Total loss: 8.8952e+10, style loss: 1.8413e+07, content loss: 7.0539e+10, time: 3.9121s Iteration: 389 Total loss: 8.8817e+10, style loss: 1.8388e+07, content loss: 7.0429e+10, time: 3.9010s Iteration: 390 Total loss: 8.8679e+10, style loss: 1.8352e+07, content loss: 7.0327e+10, time: 3.8908s Iteration: 391 Total loss: 8.8534e+10, style loss: 1.8300e+07, content loss: 7.0234e+10, time: 3.8967s Iteration: 392 Total loss: 8.8393e+10, style loss: 1.8250e+07, content loss: 7.0144e+10, time: 3.9428s Iteration: 393 Total loss: 8.8256e+10, style loss: 1.8207e+07, content loss: 7.0049e+10, time: 3.9653s Iteration: 394 Total loss: 8.8117e+10, style loss: 1.8171e+07, content loss: 6.9947e+10, time: 3.9607s Iteration: 395 Total loss: 8.7980e+10, style loss: 1.8141e+07, content loss: 6.9840e+10, time: 3.9229s Iteration: 396 Total loss: 8.7846e+10, style loss: 1.8112e+07, content loss: 6.9734e+10, time: 3.9989s Iteration: 397 Total loss: 8.7709e+10, style loss: 1.8073e+07, content loss: 6.9636e+10, time: 3.9528s Iteration: 398 Total loss: 8.7573e+10, style loss: 1.8029e+07, content loss: 6.9544e+10, time: 3.9469s Iteration: 399 Total loss: 8.7443e+10, style loss: 1.7989e+07, content loss: 6.9454e+10, time: 3.8985s Iteration: 400 Total loss: 8.7310e+10, style loss: 1.7951e+07, content loss: 6.9360e+10, time: 3.9019s Iteration: 401 Total loss: 8.7173e+10, style loss: 1.7914e+07, content loss: 6.9259e+10, time: 3.8590s Iteration: 402 Total loss: 8.7043e+10, style loss: 1.7887e+07, content loss: 6.9156e+10, time: 3.8823s Iteration: 403 Total loss: 8.6918e+10, style loss: 1.7861e+07, content loss: 6.9057e+10, time: 3.8941s Iteration: 404 Total loss: 8.6785e+10, style loss: 1.7820e+07, content loss: 6.8965e+10, time: 3.9062s Iteration: 405 Total loss: 8.6650e+10, style loss: 1.7771e+07, content loss: 6.8880e+10, time: 3.8996s Iteration: 406 Total loss: 8.6525e+10, style loss: 1.7731e+07, content loss: 6.8795e+10, time: 3.8745s Iteration: 407 Total loss: 8.6402e+10, style loss: 1.7697e+07, content loss: 6.8704e+10, time: 3.9048s Iteration: 408 Total loss: 8.6273e+10, style loss: 1.7665e+07, content loss: 6.8608e+10, time: 3.8866s Iteration: 409 Total loss: 8.6147e+10, style loss: 1.7638e+07, content loss: 6.8509e+10, time: 3.9264s Iteration: 410 Total loss: 8.6025e+10, style loss: 1.7612e+07, content loss: 6.8413e+10, time: 3.8694s Iteration: 411 Total loss: 8.5900e+10, style loss: 1.7576e+07, content loss: 6.8324e+10, time: 3.9043s Iteration: 412 Total loss: 8.5774e+10, style loss: 1.7536e+07, content loss: 6.8239e+10, time: 3.9455s Iteration: 413 Total loss: 8.5654e+10, style loss: 1.7502e+07, content loss: 6.8151e+10, time: 4.0240s Iteration: 414 Total loss: 8.5533e+10, style loss: 1.7475e+07, content loss: 6.8058e+10, time: 3.9608s Iteration: 415 Total loss: 8.5410e+10, style loss: 1.7452e+07, content loss: 6.7958e+10, time: 3.9297s Iteration: 416 Total loss: 8.5292e+10, style loss: 1.7433e+07, content loss: 6.7859e+10, time: 3.9073s Iteration: 417 Total loss: 8.5174e+10, style loss: 1.7408e+07, content loss: 6.7766e+10, time: 3.9117s Iteration: 418 Total loss: 8.5054e+10, style loss: 1.7372e+07, content loss: 6.7682e+10, time: 3.8753s Iteration: 419 Total loss: 8.4937e+10, style loss: 1.7334e+07, content loss: 6.7603e+10, time: 3.8701s Iteration: 420 Total loss: 8.4823e+10, style loss: 1.7302e+07, content loss: 6.7522e+10, time: 3.8479s Iteration: 421 Total loss: 8.4705e+10, style loss: 1.7271e+07, content loss: 6.7435e+10, time: 3.9558s Iteration: 422 Total loss: 8.4590e+10, style loss: 1.7246e+07, content loss: 6.7343e+10, time: 3.9338s Iteration: 423 Total loss: 8.4481e+10, style loss: 1.7227e+07, content loss: 6.7253e+10, time: 3.9116s Iteration: 424 Total loss: 8.4369e+10, style loss: 1.7199e+07, content loss: 6.7170e+10, time: 3.8642s Iteration: 425 Total loss: 8.4253e+10, style loss: 1.7158e+07, content loss: 6.7095e+10, time: 3.8954s Iteration: 426 Total loss: 8.4143e+10, style loss: 1.7121e+07, content loss: 6.7022e+10, time: 3.8604s Iteration: 427 Total loss: 8.4036e+10, style loss: 1.7091e+07, content loss: 6.6945e+10, time: 3.9076s Iteration: 428 Total loss: 8.3922e+10, style loss: 1.7062e+07, content loss: 6.6860e+10, time: 3.8678s Iteration: 429 Total loss: 8.3809e+10, style loss: 1.7038e+07, content loss: 6.6770e+10, time: 3.9477s Iteration: 430 Total loss: 8.3702e+10, style loss: 1.7020e+07, content loss: 6.6682e+10, time: 3.9307s Iteration: 431 Total loss: 8.3594e+10, style loss: 1.6993e+07, content loss: 6.6601e+10, time: 3.8739s Iteration: 432 Total loss: 8.3483e+10, style loss: 1.6956e+07, content loss: 6.6527e+10, time: 3.8899s Iteration: 433 Total loss: 8.3376e+10, style loss: 1.6921e+07, content loss: 6.6454e+10, time: 3.9744s Iteration: 434 Total loss: 8.3270e+10, style loss: 1.6893e+07, content loss: 6.6378e+10, time: 3.9439s Iteration: 435 Total loss: 8.3164e+10, style loss: 1.6869e+07, content loss: 6.6295e+10, time: 3.9045s Iteration: 436 Total loss: 8.3061e+10, style loss: 1.6850e+07, content loss: 6.6211e+10, time: 3.8599s Iteration: 437 Total loss: 8.2958e+10, style loss: 1.6829e+07, content loss: 6.6129e+10, time: 3.8398s Iteration: 438 Total loss: 8.2852e+10, style loss: 1.6799e+07, content loss: 6.6053e+10, time: 3.9056s Iteration: 439 Total loss: 8.2749e+10, style loss: 1.6767e+07, content loss: 6.5982e+10, time: 4.0136s Iteration: 440 Total loss: 8.2649e+10, style loss: 1.6741e+07, content loss: 6.5908e+10, time: 4.0526s Iteration: 441 Total loss: 8.2545e+10, style loss: 1.6717e+07, content loss: 6.5828e+10, time: 4.0433s Iteration: 442 Total loss: 8.2442e+10, style loss: 1.6700e+07, content loss: 6.5743e+10, time: 3.9519s Iteration: 443 Total loss: 8.2345e+10, style loss: 1.6688e+07, content loss: 6.5657e+10, time: 3.9457s Iteration: 444 Total loss: 8.2246e+10, style loss: 1.6669e+07, content loss: 6.5576e+10, time: 3.8972s Iteration: 445 Total loss: 8.2143e+10, style loss: 1.6640e+07, content loss: 6.5502e+10, time: 3.8861s Iteration: 446 Total loss: 8.2043e+10, style loss: 1.6611e+07, content loss: 6.5431e+10, time: 3.8638s Iteration: 447 Total loss: 8.1944e+10, style loss: 1.6587e+07, content loss: 6.5358e+10, time: 3.8684s Iteration: 448 Total loss: 8.1843e+10, style loss: 1.6563e+07, content loss: 6.5280e+10, time: 3.8777s Iteration: 449 Total loss: 8.1743e+10, style loss: 1.6544e+07, content loss: 6.5199e+10, time: 3.8669s Iteration: 450 Total loss: 8.1648e+10, style loss: 1.6527e+07, content loss: 6.5121e+10, time: 3.8564s Iteration: 451 Total loss: 8.1549e+10, style loss: 1.6501e+07, content loss: 6.5048e+10, time: 3.8468s Iteration: 452 Total loss: 8.1449e+10, style loss: 1.6470e+07, content loss: 6.4979e+10, time: 3.8720s Iteration: 453 Total loss: 8.1352e+10, style loss: 1.6444e+07, content loss: 6.4909e+10, time: 3.8981s Iteration: 454 Total loss: 8.1256e+10, style loss: 1.6422e+07, content loss: 6.4834e+10, time: 3.9028s Iteration: 455 Total loss: 8.1159e+10, style loss: 1.6404e+07, content loss: 6.4755e+10, time: 3.9219s Iteration: 456 Total loss: 8.1063e+10, style loss: 1.6387e+07, content loss: 6.4676e+10, time: 3.8338s Iteration: 457 Total loss: 8.0967e+10, style loss: 1.6365e+07, content loss: 6.4603e+10, time: 3.8289s Iteration: 458 Total loss: 8.0872e+10, style loss: 1.6337e+07, content loss: 6.4536e+10, time: 3.8577s Iteration: 459 Total loss: 8.0780e+10, style loss: 1.6309e+07, content loss: 6.4471e+10, time: 3.9001s Iteration: 460 Total loss: 8.0687e+10, style loss: 1.6284e+07, content loss: 6.4403e+10, time: 3.8449s Iteration: 461 Total loss: 8.0593e+10, style loss: 1.6263e+07, content loss: 6.4331e+10, time: 3.8964s Iteration: 462 Total loss: 8.0503e+10, style loss: 1.6247e+07, content loss: 6.4256e+10, time: 3.9488s Iteration: 463 Total loss: 8.0414e+10, style loss: 1.6231e+07, content loss: 6.4184e+10, time: 3.9863s Iteration: 464 Total loss: 8.0322e+10, style loss: 1.6204e+07, content loss: 6.4118e+10, time: 3.9760s Iteration: 465 Total loss: 8.0230e+10, style loss: 1.6174e+07, content loss: 6.4056e+10, time: 3.9534s Iteration: 466 Total loss: 8.0142e+10, style loss: 1.6149e+07, content loss: 6.3994e+10, time: 3.8782s Iteration: 467 Total loss: 8.0054e+10, style loss: 1.6127e+07, content loss: 6.3927e+10, time: 3.9505s Iteration: 468 Total loss: 7.9963e+10, style loss: 1.6108e+07, content loss: 6.3855e+10, time: 3.8883s Iteration: 469 Total loss: 7.9875e+10, style loss: 1.6092e+07, content loss: 6.3784e+10, time: 3.9413s Iteration: 470 Total loss: 7.9788e+10, style loss: 1.6072e+07, content loss: 6.3716e+10, time: 3.9114s Iteration: 471 Total loss: 7.9699e+10, style loss: 1.6045e+07, content loss: 6.3654e+10, time: 3.9515s Iteration: 472 Total loss: 7.9613e+10, style loss: 1.6018e+07, content loss: 6.3594e+10, time: 3.8946s Iteration: 473 Total loss: 7.9528e+10, style loss: 1.5996e+07, content loss: 6.3532e+10, time: 3.9081s Iteration: 474 Total loss: 7.9442e+10, style loss: 1.5977e+07, content loss: 6.3465e+10, time: 3.9424s Iteration: 475 Total loss: 7.9357e+10, style loss: 1.5960e+07, content loss: 6.3397e+10, time: 3.9831s Iteration: 476 Total loss: 7.9272e+10, style loss: 1.5941e+07, content loss: 6.3331e+10, time: 3.9394s Iteration: 477 Total loss: 7.9186e+10, style loss: 1.5916e+07, content loss: 6.3270e+10, time: 3.9292s Iteration: 478 Total loss: 7.9101e+10, style loss: 1.5889e+07, content loss: 6.3212e+10, time: 3.8925s Iteration: 479 Total loss: 7.9018e+10, style loss: 1.5865e+07, content loss: 6.3153e+10, time: 3.8913s Iteration: 480 Total loss: 7.8934e+10, style loss: 1.5846e+07, content loss: 6.3089e+10, time: 3.8728s Iteration: 481 Total loss: 7.8851e+10, style loss: 1.5831e+07, content loss: 6.3020e+10, time: 3.9128s Iteration: 482 Total loss: 7.8769e+10, style loss: 1.5818e+07, content loss: 6.2951e+10, time: 3.9141s Iteration: 483 Total loss: 7.8685e+10, style loss: 1.5799e+07, content loss: 6.2886e+10, time: 3.9167s Iteration: 484 Total loss: 7.8601e+10, style loss: 1.5777e+07, content loss: 6.2824e+10, time: 3.8923s Iteration: 485 Total loss: 7.8521e+10, style loss: 1.5758e+07, content loss: 6.2763e+10, time: 3.9352s Iteration: 486 Total loss: 7.8441e+10, style loss: 1.5741e+07, content loss: 6.2700e+10, time: 3.9074s Iteration: 487 Total loss: 7.8357e+10, style loss: 1.5724e+07, content loss: 6.2633e+10, time: 3.8995s Iteration: 488 Total loss: 7.8277e+10, style loss: 1.5711e+07, content loss: 6.2567e+10, time: 3.8930s Iteration: 489 Total loss: 7.8199e+10, style loss: 1.5695e+07, content loss: 6.2504e+10, time: 3.9133s Iteration: 490 Total loss: 7.8119e+10, style loss: 1.5673e+07, content loss: 6.2446e+10, time: 3.9172s Iteration: 491 Total loss: 7.8039e+10, style loss: 1.5650e+07, content loss: 6.2389e+10, time: 3.8580s Iteration: 492 Total loss: 7.7961e+10, style loss: 1.5631e+07, content loss: 6.2330e+10, time: 3.9005s Iteration: 493 Total loss: 7.7883e+10, style loss: 1.5615e+07, content loss: 6.2267e+10, time: 3.9271s Iteration: 494 Total loss: 7.7805e+10, style loss: 1.5602e+07, content loss: 6.2203e+10, time: 3.9423s Iteration: 495 Total loss: 7.7729e+10, style loss: 1.5586e+07, content loss: 6.2143e+10, time: 3.9773s Iteration: 496 Total loss: 7.7651e+10, style loss: 1.5563e+07, content loss: 6.2088e+10, time: 3.9433s Iteration: 497 Total loss: 7.7575e+10, style loss: 1.5539e+07, content loss: 6.2036e+10, time: 3.9058s Iteration: 498 Total loss: 7.7499e+10, style loss: 1.5516e+07, content loss: 6.1983e+10, time: 3.9199s Iteration: 499 Total loss: 7.7422e+10, style loss: 1.5496e+07, content loss: 6.1926e+10, time: 3.9782s Iteration: 500 Total loss: 7.7346e+10, style loss: 1.5480e+07, content loss: 6.1867e+10, time: 3.9742s Iteration: 501 Total loss: 7.7274e+10, style loss: 1.5465e+07, content loss: 6.1809e+10, time: 3.9496s Iteration: 502 Total loss: 7.7199e+10, style loss: 1.5445e+07, content loss: 6.1754e+10, time: 3.9401s Iteration: 503 Total loss: 7.7124e+10, style loss: 1.5420e+07, content loss: 6.1704e+10, time: 3.9159s Iteration: 504 Total loss: 7.7052e+10, style loss: 1.5398e+07, content loss: 6.1654e+10, time: 3.9237s Iteration: 505 Total loss: 7.6979e+10, style loss: 1.5379e+07, content loss: 6.1601e+10, time: 3.9330s Iteration: 506 Total loss: 7.6906e+10, style loss: 1.5362e+07, content loss: 6.1544e+10, time: 3.9005s Iteration: 507 Total loss: 7.6835e+10, style loss: 1.5349e+07, content loss: 6.1486e+10, time: 3.9035s Iteration: 508 Total loss: 7.6764e+10, style loss: 1.5332e+07, content loss: 6.1432e+10, time: 3.9080s Iteration: 509 Total loss: 7.6692e+10, style loss: 1.5311e+07, content loss: 6.1381e+10, time: 3.9069s Iteration: 510 Total loss: 7.6621e+10, style loss: 1.5291e+07, content loss: 6.1330e+10, time: 3.9368s Iteration: 511 Total loss: 7.6551e+10, style loss: 1.5276e+07, content loss: 6.1275e+10, time: 3.9093s Iteration: 512 Total loss: 7.6481e+10, style loss: 1.5265e+07, content loss: 6.1217e+10, time: 3.9275s Iteration: 513 Total loss: 7.6412e+10, style loss: 1.5256e+07, content loss: 6.1157e+10, time: 3.9720s Iteration: 514 Total loss: 7.6344e+10, style loss: 1.5245e+07, content loss: 6.1099e+10, time: 3.9141s Iteration: 515 Total loss: 7.6275e+10, style loss: 1.5228e+07, content loss: 6.1047e+10, time: 3.9841s Iteration: 516 Total loss: 7.6207e+10, style loss: 1.5210e+07, content loss: 6.0997e+10, time: 4.0006s Iteration: 517 Total loss: 7.6140e+10, style loss: 1.5192e+07, content loss: 6.0948e+10, time: 3.9665s Iteration: 518 Total loss: 7.6072e+10, style loss: 1.5177e+07, content loss: 6.0895e+10, time: 3.9113s Iteration: 519 Total loss: 7.6007e+10, style loss: 1.5166e+07, content loss: 6.0841e+10, time: 3.9220s Iteration: 520 Total loss: 7.5942e+10, style loss: 1.5153e+07, content loss: 6.0789e+10, time: 3.9146s Iteration: 521 Total loss: 7.5874e+10, style loss: 1.5134e+07, content loss: 6.0740e+10, time: 3.9328s Iteration: 522 Total loss: 7.5808e+10, style loss: 1.5115e+07, content loss: 6.0693e+10, time: 3.9379s Iteration: 523 Total loss: 7.5744e+10, style loss: 1.5101e+07, content loss: 6.0644e+10, time: 3.9705s Iteration: 524 Total loss: 7.5679e+10, style loss: 1.5089e+07, content loss: 6.0590e+10, time: 3.9229s Iteration: 525 Total loss: 7.5614e+10, style loss: 1.5079e+07, content loss: 6.0535e+10, time: 3.9485s Iteration: 526 Total loss: 7.5550e+10, style loss: 1.5069e+07, content loss: 6.0481e+10, time: 3.9584s Iteration: 527 Total loss: 7.5486e+10, style loss: 1.5055e+07, content loss: 6.0431e+10, time: 3.9566s Iteration: 528 Total loss: 7.5421e+10, style loss: 1.5038e+07, content loss: 6.0383e+10, time: 3.9898s Iteration: 529 Total loss: 7.5358e+10, style loss: 1.5023e+07, content loss: 6.0335e+10, time: 3.9468s Iteration: 530 Total loss: 7.5295e+10, style loss: 1.5011e+07, content loss: 6.0283e+10, time: 3.9436s Iteration: 531 Total loss: 7.5231e+10, style loss: 1.5002e+07, content loss: 6.0229e+10, time: 3.9689s Iteration: 532 Total loss: 7.5170e+10, style loss: 1.4993e+07, content loss: 6.0176e+10, time: 3.9577s Iteration: 533 Total loss: 7.5107e+10, style loss: 1.4980e+07, content loss: 6.0127e+10, time: 3.9363s Iteration: 534 Total loss: 7.5045e+10, style loss: 1.4965e+07, content loss: 6.0081e+10, time: 3.9447s Iteration: 535 Total loss: 7.4984e+10, style loss: 1.4951e+07, content loss: 6.0033e+10, time: 3.9565s Iteration: 536 Total loss: 7.4922e+10, style loss: 1.4940e+07, content loss: 5.9983e+10, time: 3.9755s Iteration: 537 Total loss: 7.4861e+10, style loss: 1.4931e+07, content loss: 5.9930e+10, time: 3.9947s Iteration: 538 Total loss: 7.4801e+10, style loss: 1.4923e+07, content loss: 5.9878e+10, time: 3.9067s Iteration: 539 Total loss: 7.4740e+10, style loss: 1.4910e+07, content loss: 5.9830e+10, time: 3.9452s Iteration: 540 Total loss: 7.4679e+10, style loss: 1.4894e+07, content loss: 5.9785e+10, time: 3.9167s Iteration: 541 Total loss: 7.4620e+10, style loss: 1.4880e+07, content loss: 5.9740e+10, time: 3.9967s Iteration: 542 Total loss: 7.4560e+10, style loss: 1.4868e+07, content loss: 5.9693e+10, time: 4.0044s Iteration: 543 Total loss: 7.4500e+10, style loss: 1.4857e+07, content loss: 5.9643e+10, time: 4.0414s Iteration: 544 Total loss: 7.4441e+10, style loss: 1.4847e+07, content loss: 5.9594e+10, time: 3.9395s Iteration: 545 Total loss: 7.4383e+10, style loss: 1.4833e+07, content loss: 5.9550e+10, time: 3.9393s Iteration: 546 Total loss: 7.4324e+10, style loss: 1.4815e+07, content loss: 5.9509e+10, time: 3.9804s Iteration: 547 Total loss: 7.4266e+10, style loss: 1.4798e+07, content loss: 5.9468e+10, time: 3.9153s Iteration: 548 Total loss: 7.4208e+10, style loss: 1.4785e+07, content loss: 5.9423e+10, time: 3.9295s Iteration: 549 Total loss: 7.4150e+10, style loss: 1.4775e+07, content loss: 5.9375e+10, time: 3.9577s Iteration: 550 Total loss: 7.4092e+10, style loss: 1.4767e+07, content loss: 5.9326e+10, time: 3.9655s Iteration: 551 Total loss: 7.4035e+10, style loss: 1.4755e+07, content loss: 5.9279e+10, time: 3.9384s Iteration: 552 Total loss: 7.3977e+10, style loss: 1.4742e+07, content loss: 5.9236e+10, time: 3.9325s Iteration: 553 Total loss: 7.3921e+10, style loss: 1.4728e+07, content loss: 5.9192e+10, time: 3.9464s Iteration: 554 Total loss: 7.3863e+10, style loss: 1.4716e+07, content loss: 5.9147e+10, time: 3.9568s Iteration: 555 Total loss: 7.3806e+10, style loss: 1.4706e+07, content loss: 5.9100e+10, time: 3.9584s Iteration: 556 Total loss: 7.3751e+10, style loss: 1.4696e+07, content loss: 5.9054e+10, time: 3.9883s Iteration: 557 Total loss: 7.3695e+10, style loss: 1.4683e+07, content loss: 5.9012e+10, time: 4.0029s Iteration: 558 Total loss: 7.3639e+10, style loss: 1.4666e+07, content loss: 5.8973e+10, time: 3.9685s Iteration: 559 Total loss: 7.3584e+10, style loss: 1.4651e+07, content loss: 5.8933e+10, time: 3.9501s Iteration: 560 Total loss: 7.3529e+10, style loss: 1.4639e+07, content loss: 5.8890e+10, time: 3.9615s Iteration: 561 Total loss: 7.3474e+10, style loss: 1.4630e+07, content loss: 5.8845e+10, time: 4.0496s Iteration: 562 Total loss: 7.3421e+10, style loss: 1.4622e+07, content loss: 5.8799e+10, time: 4.0209s Iteration: 563 Total loss: 7.3367e+10, style loss: 1.4611e+07, content loss: 5.8756e+10, time: 4.0505s Iteration: 564 Total loss: 7.3313e+10, style loss: 1.4597e+07, content loss: 5.8716e+10, time: 4.0385s Iteration: 565 Total loss: 7.3260e+10, style loss: 1.4585e+07, content loss: 5.8675e+10, time: 4.0204s Iteration: 566 Total loss: 7.3207e+10, style loss: 1.4576e+07, content loss: 5.8631e+10, time: 3.9790s Iteration: 567 Total loss: 7.3155e+10, style loss: 1.4569e+07, content loss: 5.8586e+10, time: 3.9480s Iteration: 568 Total loss: 7.3102e+10, style loss: 1.4561e+07, content loss: 5.8541e+10, time: 3.9568s Iteration: 569 Total loss: 7.3049e+10, style loss: 1.4550e+07, content loss: 5.8500e+10, time: 3.9622s Iteration: 570 Total loss: 7.2997e+10, style loss: 1.4537e+07, content loss: 5.8460e+10, time: 3.9417s Iteration: 571 Total loss: 7.2945e+10, style loss: 1.4526e+07, content loss: 5.8419e+10, time: 3.9504s Iteration: 572 Total loss: 7.2892e+10, style loss: 1.4519e+07, content loss: 5.8374e+10, time: 3.9478s Iteration: 573 Total loss: 7.2841e+10, style loss: 1.4514e+07, content loss: 5.8327e+10, time: 3.9828s Iteration: 574 Total loss: 7.2790e+10, style loss: 1.4509e+07, content loss: 5.8281e+10, time: 3.9820s Iteration: 575 Total loss: 7.2738e+10, style loss: 1.4499e+07, content loss: 5.8239e+10, time: 3.9821s Iteration: 576 Total loss: 7.2686e+10, style loss: 1.4488e+07, content loss: 5.8199e+10, time: 4.0277s Iteration: 577 Total loss: 7.2636e+10, style loss: 1.4478e+07, content loss: 5.8158e+10, time: 4.0295s Iteration: 578 Total loss: 7.2585e+10, style loss: 1.4470e+07, content loss: 5.8115e+10, time: 3.9900s Iteration: 579 Total loss: 7.2535e+10, style loss: 1.4463e+07, content loss: 5.8071e+10, time: 3.9484s Iteration: 580 Total loss: 7.2485e+10, style loss: 1.4455e+07, content loss: 5.8030e+10, time: 3.9491s Iteration: 581 Total loss: 7.2435e+10, style loss: 1.4443e+07, content loss: 5.7993e+10, time: 3.9841s Iteration: 582 Total loss: 7.2385e+10, style loss: 1.4428e+07, content loss: 5.7958e+10, time: 3.9500s Iteration: 583 Total loss: 7.2336e+10, style loss: 1.4415e+07, content loss: 5.7921e+10, time: 3.9905s Iteration: 584 Total loss: 7.2287e+10, style loss: 1.4405e+07, content loss: 5.7882e+10, time: 3.9764s Iteration: 585 Total loss: 7.2239e+10, style loss: 1.4397e+07, content loss: 5.7842e+10, time: 3.9852s Iteration: 586 Total loss: 7.2190e+10, style loss: 1.4387e+07, content loss: 5.7803e+10, time: 3.9504s Iteration: 587 Total loss: 7.2141e+10, style loss: 1.4375e+07, content loss: 5.7766e+10, time: 4.0177s Iteration: 588 Total loss: 7.2094e+10, style loss: 1.4363e+07, content loss: 5.7730e+10, time: 3.9536s Iteration: 589 Total loss: 7.2046e+10, style loss: 1.4354e+07, content loss: 5.7692e+10, time: 4.0872s Iteration: 590 Total loss: 7.1999e+10, style loss: 1.4347e+07, content loss: 5.7652e+10, time: 3.9482s Iteration: 591 Total loss: 7.1952e+10, style loss: 1.4341e+07, content loss: 5.7611e+10, time: 3.9653s Iteration: 592 Total loss: 7.1906e+10, style loss: 1.4333e+07, content loss: 5.7573e+10, time: 3.9302s Iteration: 593 Total loss: 7.1859e+10, style loss: 1.4322e+07, content loss: 5.7537e+10, time: 3.9278s Iteration: 594 Total loss: 7.1813e+10, style loss: 1.4311e+07, content loss: 5.7502e+10, time: 3.8560s Iteration: 595 Total loss: 7.1767e+10, style loss: 1.4303e+07, content loss: 5.7464e+10, time: 3.8437s Iteration: 596 Total loss: 7.1721e+10, style loss: 1.4297e+07, content loss: 5.7425e+10, time: 3.9239s Iteration: 597 Total loss: 7.1676e+10, style loss: 1.4291e+07, content loss: 5.7385e+10, time: 3.9014s Iteration: 598 Total loss: 7.1631e+10, style loss: 1.4283e+07, content loss: 5.7348e+10, time: 3.8880s Iteration: 599 Total loss: 7.1586e+10, style loss: 1.4272e+07, content loss: 5.7314e+10, time: 3.8657s Iteration: 600 Total loss: 7.1541e+10, style loss: 1.4262e+07, content loss: 5.7279e+10, time: 3.8686s Iteration: 601 Total loss: 7.1496e+10, style loss: 1.4254e+07, content loss: 5.7242e+10, time: 3.8598s Iteration: 602 Total loss: 7.1451e+10, style loss: 1.4247e+07, content loss: 5.7205e+10, time: 3.8654s Iteration: 603 Total loss: 7.1407e+10, style loss: 1.4239e+07, content loss: 5.7168e+10, time: 3.8470s Iteration: 604 Total loss: 7.1363e+10, style loss: 1.4227e+07, content loss: 5.7135e+10, time: 3.9015s Iteration: 605 Total loss: 7.1319e+10, style loss: 1.4215e+07, content loss: 5.7104e+10, time: 3.8451s Iteration: 606 Total loss: 7.1276e+10, style loss: 1.4204e+07, content loss: 5.7072e+10, time: 3.8963s Iteration: 607 Total loss: 7.1232e+10, style loss: 1.4194e+07, content loss: 5.7038e+10, time: 3.8691s Iteration: 608 Total loss: 7.1189e+10, style loss: 1.4186e+07, content loss: 5.7002e+10, time: 3.8696s Iteration: 609 Total loss: 7.1146e+10, style loss: 1.4178e+07, content loss: 5.6968e+10, time: 3.8434s Iteration: 610 Total loss: 7.1103e+10, style loss: 1.4166e+07, content loss: 5.6937e+10, time: 3.8512s Iteration: 611 Total loss: 7.1060e+10, style loss: 1.4154e+07, content loss: 5.6906e+10, time: 3.8732s Iteration: 612 Total loss: 7.1019e+10, style loss: 1.4145e+07, content loss: 5.6873e+10, time: 3.8596s Iteration: 613 Total loss: 7.0976e+10, style loss: 1.4138e+07, content loss: 5.6838e+10, time: 3.8632s Iteration: 614 Total loss: 7.0934e+10, style loss: 1.4132e+07, content loss: 5.6802e+10, time: 3.8745s Iteration: 615 Total loss: 7.0893e+10, style loss: 1.4124e+07, content loss: 5.6769e+10, time: 3.9052s Iteration: 616 Total loss: 7.0851e+10, style loss: 1.4113e+07, content loss: 5.6738e+10, time: 3.8582s Iteration: 617 Total loss: 7.0809e+10, style loss: 1.4102e+07, content loss: 5.6708e+10, time: 3.9264s Iteration: 618 Total loss: 7.0768e+10, style loss: 1.4093e+07, content loss: 5.6675e+10, time: 3.9026s Iteration: 619 Total loss: 7.0727e+10, style loss: 1.4086e+07, content loss: 5.6641e+10, time: 3.8677s Iteration: 620 Total loss: 7.0686e+10, style loss: 1.4078e+07, content loss: 5.6607e+10, time: 3.8983s Iteration: 621 Total loss: 7.0645e+10, style loss: 1.4069e+07, content loss: 5.6576e+10, time: 3.9382s Iteration: 622 Total loss: 7.0605e+10, style loss: 1.4059e+07, content loss: 5.6546e+10, time: 3.9420s Iteration: 623 Total loss: 7.0565e+10, style loss: 1.4050e+07, content loss: 5.6515e+10, time: 3.8598s Iteration: 624 Total loss: 7.0524e+10, style loss: 1.4042e+07, content loss: 5.6481e+10, time: 3.8469s Iteration: 625 Total loss: 7.0484e+10, style loss: 1.4037e+07, content loss: 5.6447e+10, time: 3.8597s Iteration: 626 Total loss: 7.0445e+10, style loss: 1.4031e+07, content loss: 5.6414e+10, time: 3.8413s Iteration: 627 Total loss: 7.0405e+10, style loss: 1.4022e+07, content loss: 5.6383e+10, time: 3.8913s Iteration: 628 Total loss: 7.0366e+10, style loss: 1.4013e+07, content loss: 5.6353e+10, time: 3.8597s Iteration: 629 Total loss: 7.0327e+10, style loss: 1.4006e+07, content loss: 5.6321e+10, time: 3.8639s Iteration: 630 Total loss: 7.0287e+10, style loss: 1.4001e+07, content loss: 5.6287e+10, time: 3.8582s Iteration: 631 Total loss: 7.0248e+10, style loss: 1.3996e+07, content loss: 5.6252e+10, time: 3.8814s Iteration: 632 Total loss: 7.0210e+10, style loss: 1.3990e+07, content loss: 5.6220e+10, time: 3.8413s Iteration: 633 Total loss: 7.0171e+10, style loss: 1.3982e+07, content loss: 5.6189e+10, time: 3.8401s Iteration: 634 Total loss: 7.0133e+10, style loss: 1.3974e+07, content loss: 5.6159e+10, time: 3.8102s Iteration: 635 Total loss: 7.0095e+10, style loss: 1.3967e+07, content loss: 5.6128e+10, time: 3.8716s Iteration: 636 Total loss: 7.0057e+10, style loss: 1.3962e+07, content loss: 5.6095e+10, time: 3.8228s Iteration: 637 Total loss: 7.0019e+10, style loss: 1.3955e+07, content loss: 5.6064e+10, time: 3.8862s Iteration: 638 Total loss: 6.9980e+10, style loss: 1.3945e+07, content loss: 5.6035e+10, time: 3.9279s Iteration: 639 Total loss: 6.9943e+10, style loss: 1.3935e+07, content loss: 5.6008e+10, time: 3.9030s Iteration: 640 Total loss: 6.9905e+10, style loss: 1.3926e+07, content loss: 5.5980e+10, time: 3.8397s Iteration: 641 Total loss: 6.9868e+10, style loss: 1.3919e+07, content loss: 5.5949e+10, time: 3.8493s Iteration: 642 Total loss: 6.9831e+10, style loss: 1.3912e+07, content loss: 5.5918e+10, time: 3.8398s Iteration: 643 Total loss: 6.9794e+10, style loss: 1.3904e+07, content loss: 5.5889e+10, time: 3.8419s Iteration: 644 Total loss: 6.9756e+10, style loss: 1.3893e+07, content loss: 5.5863e+10, time: 3.8493s Iteration: 645 Total loss: 6.9720e+10, style loss: 1.3882e+07, content loss: 5.5837e+10, time: 3.8401s Iteration: 646 Total loss: 6.9683e+10, style loss: 1.3874e+07, content loss: 5.5809e+10, time: 3.8747s Iteration: 647 Total loss: 6.9647e+10, style loss: 1.3867e+07, content loss: 5.5779e+10, time: 3.8399s Iteration: 648 Total loss: 6.9610e+10, style loss: 1.3861e+07, content loss: 5.5750e+10, time: 3.8579s Iteration: 649 Total loss: 6.9574e+10, style loss: 1.3853e+07, content loss: 5.5721e+10, time: 3.8295s Iteration: 650 Total loss: 6.9538e+10, style loss: 1.3844e+07, content loss: 5.5694e+10, time: 3.8512s Iteration: 651 Total loss: 6.9503e+10, style loss: 1.3837e+07, content loss: 5.5665e+10, time: 3.9841s Iteration: 652 Total loss: 6.9467e+10, style loss: 1.3832e+07, content loss: 5.5635e+10, time: 3.8470s Iteration: 653 Total loss: 6.9432e+10, style loss: 1.3828e+07, content loss: 5.5604e+10, time: 3.8716s Iteration: 654 Total loss: 6.9397e+10, style loss: 1.3822e+07, content loss: 5.5575e+10, time: 3.8452s Iteration: 655 Total loss: 6.9362e+10, style loss: 1.3814e+07, content loss: 5.5548e+10, time: 3.8826s Iteration: 656 Total loss: 6.9328e+10, style loss: 1.3807e+07, content loss: 5.5521e+10, time: 3.8853s Iteration: 657 Total loss: 6.9293e+10, style loss: 1.3801e+07, content loss: 5.5492e+10, time: 3.8599s Iteration: 658 Total loss: 6.9259e+10, style loss: 1.3797e+07, content loss: 5.5462e+10, time: 3.9222s Iteration: 659 Total loss: 6.9225e+10, style loss: 1.3792e+07, content loss: 5.5432e+10, time: 3.8901s Iteration: 660 Total loss: 6.9190e+10, style loss: 1.3786e+07, content loss: 5.5405e+10, time: 3.9035s Iteration: 661 Total loss: 6.9156e+10, style loss: 1.3778e+07, content loss: 5.5378e+10, time: 3.8541s Iteration: 662 Total loss: 6.9122e+10, style loss: 1.3770e+07, content loss: 5.5352e+10, time: 3.8479s Iteration: 663 Total loss: 6.9088e+10, style loss: 1.3764e+07, content loss: 5.5324e+10, time: 3.8269s Iteration: 664 Total loss: 6.9055e+10, style loss: 1.3759e+07, content loss: 5.5296e+10, time: 3.8462s Iteration: 665 Total loss: 6.9021e+10, style loss: 1.3753e+07, content loss: 5.5268e+10, time: 3.8163s Iteration: 666 Total loss: 6.8988e+10, style loss: 1.3745e+07, content loss: 5.5243e+10, time: 3.8549s Iteration: 667 Total loss: 6.8955e+10, style loss: 1.3737e+07, content loss: 5.5218e+10, time: 3.8313s Iteration: 668 Total loss: 6.8921e+10, style loss: 1.3730e+07, content loss: 5.5191e+10, time: 3.8572s Iteration: 669 Total loss: 6.8888e+10, style loss: 1.3724e+07, content loss: 5.5164e+10, time: 3.8683s Iteration: 670 Total loss: 6.8855e+10, style loss: 1.3718e+07, content loss: 5.5138e+10, time: 3.9045s Iteration: 671 Total loss: 6.8822e+10, style loss: 1.3709e+07, content loss: 5.5113e+10, time: 3.8604s Iteration: 672 Total loss: 6.8789e+10, style loss: 1.3700e+07, content loss: 5.5089e+10, time: 3.8608s Iteration: 673 Total loss: 6.8757e+10, style loss: 1.3692e+07, content loss: 5.5064e+10, time: 3.8602s Iteration: 674 Total loss: 6.8724e+10, style loss: 1.3686e+07, content loss: 5.5038e+10, time: 3.8273s Iteration: 675 Total loss: 6.8691e+10, style loss: 1.3681e+07, content loss: 5.5011e+10, time: 3.8159s Iteration: 676 Total loss: 6.8659e+10, style loss: 1.3674e+07, content loss: 5.4985e+10, time: 3.8582s Iteration: 677 Total loss: 6.8627e+10, style loss: 1.3666e+07, content loss: 5.4961e+10, time: 3.8459s Iteration: 678 Total loss: 6.8595e+10, style loss: 1.3657e+07, content loss: 5.4938e+10, time: 3.8544s Iteration: 679 Total loss: 6.8563e+10, style loss: 1.3649e+07, content loss: 5.4914e+10, time: 3.9098s Iteration: 680 Total loss: 6.8531e+10, style loss: 1.3643e+07, content loss: 5.4888e+10, time: 3.8962s Iteration: 681 Total loss: 6.8499e+10, style loss: 1.3636e+07, content loss: 5.4863e+10, time: 3.8992s Iteration: 682 Total loss: 6.8468e+10, style loss: 1.3629e+07, content loss: 5.4839e+10, time: 3.8250s Iteration: 683 Total loss: 6.8436e+10, style loss: 1.3621e+07, content loss: 5.4815e+10, time: 3.8168s Iteration: 684 Total loss: 6.8405e+10, style loss: 1.3615e+07, content loss: 5.4790e+10, time: 3.8534s Iteration: 685 Total loss: 6.8373e+10, style loss: 1.3611e+07, content loss: 5.4763e+10, time: 3.8544s Iteration: 686 Total loss: 6.8343e+10, style loss: 1.3607e+07, content loss: 5.4736e+10, time: 3.8504s Iteration: 687 Total loss: 6.8312e+10, style loss: 1.3601e+07, content loss: 5.4711e+10, time: 3.8381s Iteration: 688 Total loss: 6.8281e+10, style loss: 1.3593e+07, content loss: 5.4688e+10, time: 3.8502s Iteration: 689 Total loss: 6.8250e+10, style loss: 1.3585e+07, content loss: 5.4666e+10, time: 3.8581s Iteration: 690 Total loss: 6.8220e+10, style loss: 1.3577e+07, content loss: 5.4642e+10, time: 3.8071s Iteration: 691 Total loss: 6.8189e+10, style loss: 1.3571e+07, content loss: 5.4618e+10, time: 3.8337s Iteration: 692 Total loss: 6.8159e+10, style loss: 1.3564e+07, content loss: 5.4595e+10, time: 3.8273s Iteration: 693 Total loss: 6.8129e+10, style loss: 1.3557e+07, content loss: 5.4573e+10, time: 3.8525s Iteration: 694 Total loss: 6.8099e+10, style loss: 1.3549e+07, content loss: 5.4550e+10, time: 3.8269s Iteration: 695 Total loss: 6.8070e+10, style loss: 1.3543e+07, content loss: 5.4527e+10, time: 3.8408s Iteration: 696 Total loss: 6.8040e+10, style loss: 1.3538e+07, content loss: 5.4503e+10, time: 3.8402s Iteration: 697 Total loss: 6.8011e+10, style loss: 1.3532e+07, content loss: 5.4479e+10, time: 3.8979s Iteration: 698 Total loss: 6.7982e+10, style loss: 1.3524e+07, content loss: 5.4458e+10, time: 3.8215s Iteration: 699 Total loss: 6.7952e+10, style loss: 1.3515e+07, content loss: 5.4437e+10, time: 3.8502s Iteration: 700 Total loss: 6.7924e+10, style loss: 1.3507e+07, content loss: 5.4416e+10, time: 3.9382s Iteration: 701 Total loss: 6.7895e+10, style loss: 1.3501e+07, content loss: 5.4394e+10, time: 4.0843s Iteration: 702 Total loss: 6.7866e+10, style loss: 1.3495e+07, content loss: 5.4371e+10, time: 4.0181s Iteration: 703 Total loss: 6.7838e+10, style loss: 1.3488e+07, content loss: 5.4350e+10, time: 3.9675s Iteration: 704 Total loss: 6.7809e+10, style loss: 1.3480e+07, content loss: 5.4329e+10, time: 3.9219s Iteration: 705 Total loss: 6.7781e+10, style loss: 1.3472e+07, content loss: 5.4309e+10, time: 3.9201s Iteration: 706 Total loss: 6.7752e+10, style loss: 1.3466e+07, content loss: 5.4287e+10, time: 3.8328s Iteration: 707 Total loss: 6.7724e+10, style loss: 1.3460e+07, content loss: 5.4264e+10, time: 4.1380s Iteration: 708 Total loss: 6.7696e+10, style loss: 1.3454e+07, content loss: 5.4243e+10, time: 4.1909s Iteration: 709 Total loss: 6.7668e+10, style loss: 1.3446e+07, content loss: 5.4222e+10, time: 4.1336s Iteration: 710 Total loss: 6.7640e+10, style loss: 1.3438e+07, content loss: 5.4202e+10, time: 3.9856s Iteration: 711 Total loss: 6.7613e+10, style loss: 1.3432e+07, content loss: 5.4181e+10, time: 4.0158s Iteration: 712 Total loss: 6.7585e+10, style loss: 1.3425e+07, content loss: 5.4160e+10, time: 4.1082s Iteration: 713 Total loss: 6.7557e+10, style loss: 1.3419e+07, content loss: 5.4138e+10, time: 4.1775s Iteration: 714 Total loss: 6.7530e+10, style loss: 1.3411e+07, content loss: 5.4118e+10, time: 4.1597s Iteration: 715 Total loss: 6.7502e+10, style loss: 1.3404e+07, content loss: 5.4099e+10, time: 4.1737s Iteration: 716 Total loss: 6.7475e+10, style loss: 1.3397e+07, content loss: 5.4078e+10, time: 4.1444s Iteration: 717 Total loss: 6.7448e+10, style loss: 1.3392e+07, content loss: 5.4056e+10, time: 4.1685s Iteration: 718 Total loss: 6.7421e+10, style loss: 1.3388e+07, content loss: 5.4034e+10, time: 4.1239s Iteration: 719 Total loss: 6.7394e+10, style loss: 1.3382e+07, content loss: 5.4012e+10, time: 3.9834s Iteration: 720 Total loss: 6.7367e+10, style loss: 1.3375e+07, content loss: 5.3992e+10, time: 4.0465s Iteration: 721 Total loss: 6.7341e+10, style loss: 1.3369e+07, content loss: 5.3972e+10, time: 3.9994s Iteration: 722 Total loss: 6.7314e+10, style loss: 1.3364e+07, content loss: 5.3950e+10, time: 3.9734s Iteration: 723 Total loss: 6.7287e+10, style loss: 1.3359e+07, content loss: 5.3928e+10, time: 3.9740s Iteration: 724 Total loss: 6.7261e+10, style loss: 1.3354e+07, content loss: 5.3907e+10, time: 3.9373s Iteration: 725 Total loss: 6.7234e+10, style loss: 1.3348e+07, content loss: 5.3886e+10, time: 3.9843s Iteration: 726 Total loss: 6.7208e+10, style loss: 1.3343e+07, content loss: 5.3865e+10, time: 3.9477s Iteration: 727 Total loss: 6.7182e+10, style loss: 1.3339e+07, content loss: 5.3843e+10, time: 4.0241s Iteration: 728 Total loss: 6.7155e+10, style loss: 1.3335e+07, content loss: 5.3820e+10, time: 3.9520s Iteration: 729 Total loss: 6.7129e+10, style loss: 1.3331e+07, content loss: 5.3798e+10, time: 4.0171s Iteration: 730 Total loss: 6.7103e+10, style loss: 1.3326e+07, content loss: 5.3777e+10, time: 3.9819s Iteration: 731 Total loss: 6.7077e+10, style loss: 1.3321e+07, content loss: 5.3756e+10, time: 3.9337s Iteration: 732 Total loss: 6.7051e+10, style loss: 1.3317e+07, content loss: 5.3733e+10, time: 3.9844s Iteration: 733 Total loss: 6.7025e+10, style loss: 1.3314e+07, content loss: 5.3710e+10, time: 4.0278s Iteration: 734 Total loss: 6.6999e+10, style loss: 1.3311e+07, content loss: 5.3688e+10, time: 3.9958s Iteration: 735 Total loss: 6.6973e+10, style loss: 1.3306e+07, content loss: 5.3667e+10, time: 4.0051s Iteration: 736 Total loss: 6.6947e+10, style loss: 1.3301e+07, content loss: 5.3647e+10, time: 3.9844s Iteration: 737 Total loss: 6.6922e+10, style loss: 1.3296e+07, content loss: 5.3626e+10, time: 3.9717s Iteration: 738 Total loss: 6.6896e+10, style loss: 1.3292e+07, content loss: 5.3604e+10, time: 4.0135s Iteration: 739 Total loss: 6.6871e+10, style loss: 1.3288e+07, content loss: 5.3583e+10, time: 3.9722s Iteration: 740 Total loss: 6.6845e+10, style loss: 1.3283e+07, content loss: 5.3562e+10, time: 4.0010s Iteration: 741 Total loss: 6.6820e+10, style loss: 1.3278e+07, content loss: 5.3542e+10, time: 3.9975s Iteration: 742 Total loss: 6.6795e+10, style loss: 1.3273e+07, content loss: 5.3522e+10, time: 4.0109s Iteration: 743 Total loss: 6.6770e+10, style loss: 1.3270e+07, content loss: 5.3500e+10, time: 3.9610s Iteration: 744 Total loss: 6.6745e+10, style loss: 1.3266e+07, content loss: 5.3478e+10, time: 4.0260s Iteration: 745 Total loss: 6.6720e+10, style loss: 1.3262e+07, content loss: 5.3458e+10, time: 3.9400s Iteration: 746 Total loss: 6.6695e+10, style loss: 1.3257e+07, content loss: 5.3438e+10, time: 3.9527s Iteration: 747 Total loss: 6.6671e+10, style loss: 1.3252e+07, content loss: 5.3419e+10, time: 3.9512s Iteration: 748 Total loss: 6.6646e+10, style loss: 1.3248e+07, content loss: 5.3398e+10, time: 3.9852s Iteration: 749 Total loss: 6.6622e+10, style loss: 1.3244e+07, content loss: 5.3378e+10, time: 3.9790s Iteration: 750 Total loss: 6.6598e+10, style loss: 1.3240e+07, content loss: 5.3358e+10, time: 4.0099s Iteration: 751 Total loss: 6.6573e+10, style loss: 1.3235e+07, content loss: 5.3339e+10, time: 3.9662s Iteration: 752 Total loss: 6.6549e+10, style loss: 1.3230e+07, content loss: 5.3320e+10, time: 3.9705s Iteration: 753 Total loss: 6.6525e+10, style loss: 1.3226e+07, content loss: 5.3300e+10, time: 3.9910s Iteration: 754 Total loss: 6.6501e+10, style loss: 1.3222e+07, content loss: 5.3279e+10, time: 3.9459s Iteration: 755 Total loss: 6.6477e+10, style loss: 1.3218e+07, content loss: 5.3259e+10, time: 3.9377s Iteration: 756 Total loss: 6.6454e+10, style loss: 1.3213e+07, content loss: 5.3240e+10, time: 3.9914s Iteration: 757 Total loss: 6.6430e+10, style loss: 1.3209e+07, content loss: 5.3221e+10, time: 4.0092s Iteration: 758 Total loss: 6.6406e+10, style loss: 1.3204e+07, content loss: 5.3202e+10, time: 3.9580s Iteration: 759 Total loss: 6.6382e+10, style loss: 1.3201e+07, content loss: 5.3182e+10, time: 3.9994s Iteration: 760 Total loss: 6.6359e+10, style loss: 1.3197e+07, content loss: 5.3162e+10, time: 3.9972s Iteration: 761 Total loss: 6.6336e+10, style loss: 1.3191e+07, content loss: 5.3144e+10, time: 4.0062s Iteration: 762 Total loss: 6.6312e+10, style loss: 1.3186e+07, content loss: 5.3126e+10, time: 4.0221s Iteration: 763 Total loss: 6.6289e+10, style loss: 1.3181e+07, content loss: 5.3108e+10, time: 3.9774s Iteration: 764 Total loss: 6.6266e+10, style loss: 1.3177e+07, content loss: 5.3088e+10, time: 3.9552s Iteration: 765 Total loss: 6.6243e+10, style loss: 1.3174e+07, content loss: 5.3069e+10, time: 3.9802s Iteration: 766 Total loss: 6.6220e+10, style loss: 1.3169e+07, content loss: 5.3051e+10, time: 3.9711s Iteration: 767 Total loss: 6.6197e+10, style loss: 1.3164e+07, content loss: 5.3033e+10, time: 3.9648s Iteration: 768 Total loss: 6.6174e+10, style loss: 1.3160e+07, content loss: 5.3014e+10, time: 4.0051s Iteration: 769 Total loss: 6.6152e+10, style loss: 1.3156e+07, content loss: 5.2996e+10, time: 3.9548s Iteration: 770 Total loss: 6.6129e+10, style loss: 1.3151e+07, content loss: 5.2978e+10, time: 3.9464s Iteration: 771 Total loss: 6.6106e+10, style loss: 1.3145e+07, content loss: 5.2961e+10, time: 3.9659s Iteration: 772 Total loss: 6.6084e+10, style loss: 1.3139e+07, content loss: 5.2945e+10, time: 3.9478s Iteration: 773 Total loss: 6.6062e+10, style loss: 1.3134e+07, content loss: 5.2928e+10, time: 3.9177s Iteration: 774 Total loss: 6.6039e+10, style loss: 1.3129e+07, content loss: 5.2910e+10, time: 3.9706s Iteration: 775 Total loss: 6.6017e+10, style loss: 1.3125e+07, content loss: 5.2892e+10, time: 3.9508s Iteration: 776 Total loss: 6.5995e+10, style loss: 1.3120e+07, content loss: 5.2875e+10, time: 3.9212s Iteration: 777 Total loss: 6.5974e+10, style loss: 1.3115e+07, content loss: 5.2859e+10, time: 3.9818s Iteration: 778 Total loss: 6.5952e+10, style loss: 1.3110e+07, content loss: 5.2842e+10, time: 4.0070s Iteration: 779 Total loss: 6.5930e+10, style loss: 1.3106e+07, content loss: 5.2824e+10, time: 4.0261s Iteration: 780 Total loss: 6.5908e+10, style loss: 1.3101e+07, content loss: 5.2807e+10, time: 4.0521s Iteration: 781 Total loss: 6.5887e+10, style loss: 1.3096e+07, content loss: 5.2791e+10, time: 4.0071s Iteration: 782 Total loss: 6.5865e+10, style loss: 1.3090e+07, content loss: 5.2775e+10, time: 4.0172s Iteration: 783 Total loss: 6.5844e+10, style loss: 1.3086e+07, content loss: 5.2758e+10, time: 4.0280s Iteration: 784 Total loss: 6.5822e+10, style loss: 1.3082e+07, content loss: 5.2741e+10, time: 3.9432s Iteration: 785 Total loss: 6.5801e+10, style loss: 1.3078e+07, content loss: 5.2723e+10, time: 3.9580s Iteration: 786 Total loss: 6.5780e+10, style loss: 1.3072e+07, content loss: 5.2708e+10, time: 3.9904s Iteration: 787 Total loss: 6.5759e+10, style loss: 1.3066e+07, content loss: 5.2692e+10, time: 3.9760s Iteration: 788 Total loss: 6.5738e+10, style loss: 1.3061e+07, content loss: 5.2676e+10, time: 3.9286s Iteration: 789 Total loss: 6.5716e+10, style loss: 1.3057e+07, content loss: 5.2660e+10, time: 3.9795s Iteration: 790 Total loss: 6.5696e+10, style loss: 1.3052e+07, content loss: 5.2643e+10, time: 3.9672s Iteration: 791 Total loss: 6.5675e+10, style loss: 1.3046e+07, content loss: 5.2628e+10, time: 3.9531s Iteration: 792 Total loss: 6.5654e+10, style loss: 1.3039e+07, content loss: 5.2614e+10, time: 3.9531s Iteration: 793 Total loss: 6.5633e+10, style loss: 1.3033e+07, content loss: 5.2600e+10, time: 3.9634s Iteration: 794 Total loss: 6.5612e+10, style loss: 1.3027e+07, content loss: 5.2585e+10, time: 3.9795s Iteration: 795 Total loss: 6.5592e+10, style loss: 1.3022e+07, content loss: 5.2569e+10, time: 3.9639s Iteration: 796 Total loss: 6.5571e+10, style loss: 1.3017e+07, content loss: 5.2554e+10, time: 3.9425s Iteration: 797 Total loss: 6.5551e+10, style loss: 1.3012e+07, content loss: 5.2538e+10, time: 3.9692s Iteration: 798 Total loss: 6.5530e+10, style loss: 1.3009e+07, content loss: 5.2522e+10, time: 4.0013s Iteration: 799 Total loss: 6.5510e+10, style loss: 1.3006e+07, content loss: 5.2504e+10, time: 3.9454s Iteration: 800 Total loss: 6.5490e+10, style loss: 1.3003e+07, content loss: 5.2487e+10, time: 3.9507s Iteration: 801 Total loss: 6.5470e+10, style loss: 1.2998e+07, content loss: 5.2471e+10, time: 3.9748s Iteration: 802 Total loss: 6.5450e+10, style loss: 1.2993e+07, content loss: 5.2457e+10, time: 3.9719s Iteration: 803 Total loss: 6.5430e+10, style loss: 1.2988e+07, content loss: 5.2442e+10, time: 3.9319s Iteration: 804 Total loss: 6.5410e+10, style loss: 1.2983e+07, content loss: 5.2427e+10, time: 3.9483s Iteration: 805 Total loss: 6.5390e+10, style loss: 1.2979e+07, content loss: 5.2411e+10, time: 3.9505s Iteration: 806 Total loss: 6.5370e+10, style loss: 1.2974e+07, content loss: 5.2396e+10, time: 3.9425s Iteration: 807 Total loss: 6.5351e+10, style loss: 1.2970e+07, content loss: 5.2381e+10, time: 3.9338s Iteration: 808 Total loss: 6.5331e+10, style loss: 1.2966e+07, content loss: 5.2365e+10, time: 3.9251s Iteration: 809 Total loss: 6.5312e+10, style loss: 1.2963e+07, content loss: 5.2349e+10, time: 3.9514s Iteration: 810 Total loss: 6.5293e+10, style loss: 1.2959e+07, content loss: 5.2334e+10, time: 4.0493s Iteration: 811 Total loss: 6.5273e+10, style loss: 1.2954e+07, content loss: 5.2319e+10, time: 3.9494s Iteration: 812 Total loss: 6.5254e+10, style loss: 1.2949e+07, content loss: 5.2305e+10, time: 3.9925s Iteration: 813 Total loss: 6.5235e+10, style loss: 1.2945e+07, content loss: 5.2290e+10, time: 4.0165s Iteration: 814 Total loss: 6.5216e+10, style loss: 1.2942e+07, content loss: 5.2274e+10, time: 3.9799s Iteration: 815 Total loss: 6.5197e+10, style loss: 1.2938e+07, content loss: 5.2259e+10, time: 3.9665s Iteration: 816 Total loss: 6.5178e+10, style loss: 1.2934e+07, content loss: 5.2244e+10, time: 3.9532s Iteration: 817 Total loss: 6.5160e+10, style loss: 1.2929e+07, content loss: 5.2230e+10, time: 3.9310s Iteration: 818 Total loss: 6.5141e+10, style loss: 1.2925e+07, content loss: 5.2216e+10, time: 3.9531s Iteration: 819 Total loss: 6.5123e+10, style loss: 1.2922e+07, content loss: 5.2201e+10, time: 3.9128s Iteration: 820 Total loss: 6.5104e+10, style loss: 1.2918e+07, content loss: 5.2186e+10, time: 3.9763s Iteration: 821 Total loss: 6.5086e+10, style loss: 1.2914e+07, content loss: 5.2172e+10, time: 4.0197s Iteration: 822 Total loss: 6.5068e+10, style loss: 1.2909e+07, content loss: 5.2159e+10, time: 4.0094s Iteration: 823 Total loss: 6.5050e+10, style loss: 1.2905e+07, content loss: 5.2145e+10, time: 3.9467s Iteration: 824 Total loss: 6.5032e+10, style loss: 1.2901e+07, content loss: 5.2131e+10, time: 3.9533s Iteration: 825 Total loss: 6.5014e+10, style loss: 1.2896e+07, content loss: 5.2117e+10, time: 3.9777s Iteration: 826 Total loss: 6.4996e+10, style loss: 1.2891e+07, content loss: 5.2104e+10, time: 3.9531s Iteration: 827 Total loss: 6.4978e+10, style loss: 1.2886e+07, content loss: 5.2092e+10, time: 3.9534s Iteration: 828 Total loss: 6.4960e+10, style loss: 1.2881e+07, content loss: 5.2079e+10, time: 3.9707s Iteration: 829 Total loss: 6.4942e+10, style loss: 1.2877e+07, content loss: 5.2064e+10, time: 3.9585s Iteration: 830 Total loss: 6.4924e+10, style loss: 1.2874e+07, content loss: 5.2050e+10, time: 3.9746s Iteration: 831 Total loss: 6.4906e+10, style loss: 1.2870e+07, content loss: 5.2036e+10, time: 3.9454s Iteration: 832 Total loss: 6.4889e+10, style loss: 1.2866e+07, content loss: 5.2023e+10, time: 3.9052s Iteration: 833 Total loss: 6.4871e+10, style loss: 1.2863e+07, content loss: 5.2008e+10, time: 3.9308s Iteration: 834 Total loss: 6.4853e+10, style loss: 1.2860e+07, content loss: 5.1994e+10, time: 3.9365s Iteration: 835 Total loss: 6.4836e+10, style loss: 1.2856e+07, content loss: 5.1980e+10, time: 3.9293s Iteration: 836 Total loss: 6.4818e+10, style loss: 1.2852e+07, content loss: 5.1966e+10, time: 3.9627s Iteration: 837 Total loss: 6.4801e+10, style loss: 1.2848e+07, content loss: 5.1953e+10, time: 3.9437s Iteration: 838 Total loss: 6.4784e+10, style loss: 1.2844e+07, content loss: 5.1940e+10, time: 3.9147s Iteration: 839 Total loss: 6.4766e+10, style loss: 1.2840e+07, content loss: 5.1926e+10, time: 3.9466s Iteration: 840 Total loss: 6.4749e+10, style loss: 1.2836e+07, content loss: 5.1913e+10, time: 3.9554s Iteration: 841 Total loss: 6.4732e+10, style loss: 1.2831e+07, content loss: 5.1901e+10, time: 3.9479s Iteration: 842 Total loss: 6.4715e+10, style loss: 1.2826e+07, content loss: 5.1889e+10, time: 4.0000s Iteration: 843 Total loss: 6.4698e+10, style loss: 1.2821e+07, content loss: 5.1876e+10, time: 3.9712s Iteration: 844 Total loss: 6.4681e+10, style loss: 1.2817e+07, content loss: 5.1864e+10, time: 3.9188s Iteration: 845 Total loss: 6.4664e+10, style loss: 1.2813e+07, content loss: 5.1851e+10, time: 3.9503s Iteration: 846 Total loss: 6.4647e+10, style loss: 1.2807e+07, content loss: 5.1840e+10, time: 3.9482s Iteration: 847 Total loss: 6.4630e+10, style loss: 1.2802e+07, content loss: 5.1828e+10, time: 3.8797s Iteration: 848 Total loss: 6.4613e+10, style loss: 1.2798e+07, content loss: 5.1816e+10, time: 3.9291s Iteration: 849 Total loss: 6.4597e+10, style loss: 1.2794e+07, content loss: 5.1803e+10, time: 3.9229s Iteration: 850 Total loss: 6.4580e+10, style loss: 1.2789e+07, content loss: 5.1790e+10, time: 3.9172s Iteration: 851 Total loss: 6.4563e+10, style loss: 1.2785e+07, content loss: 5.1778e+10, time: 3.9241s Iteration: 852 Total loss: 6.4547e+10, style loss: 1.2782e+07, content loss: 5.1765e+10, time: 3.9644s Iteration: 853 Total loss: 6.4530e+10, style loss: 1.2779e+07, content loss: 5.1751e+10, time: 3.9369s Iteration: 854 Total loss: 6.4514e+10, style loss: 1.2776e+07, content loss: 5.1737e+10, time: 3.9303s Iteration: 855 Total loss: 6.4497e+10, style loss: 1.2773e+07, content loss: 5.1724e+10, time: 3.9364s Iteration: 856 Total loss: 6.4481e+10, style loss: 1.2770e+07, content loss: 5.1711e+10, time: 3.9629s Iteration: 857 Total loss: 6.4464e+10, style loss: 1.2766e+07, content loss: 5.1698e+10, time: 4.0116s Iteration: 858 Total loss: 6.4448e+10, style loss: 1.2763e+07, content loss: 5.1685e+10, time: 4.0118s Iteration: 859 Total loss: 6.4432e+10, style loss: 1.2760e+07, content loss: 5.1671e+10, time: 3.9572s Iteration: 860 Total loss: 6.4416e+10, style loss: 1.2758e+07, content loss: 5.1658e+10, time: 3.9376s Iteration: 861 Total loss: 6.4399e+10, style loss: 1.2754e+07, content loss: 5.1645e+10, time: 3.9765s Iteration: 862 Total loss: 6.4383e+10, style loss: 1.2752e+07, content loss: 5.1632e+10, time: 3.9569s Iteration: 863 Total loss: 6.4367e+10, style loss: 1.2749e+07, content loss: 5.1618e+10, time: 3.9903s Iteration: 864 Total loss: 6.4351e+10, style loss: 1.2746e+07, content loss: 5.1605e+10, time: 3.9198s Iteration: 865 Total loss: 6.4335e+10, style loss: 1.2743e+07, content loss: 5.1593e+10, time: 3.9069s Iteration: 866 Total loss: 6.4319e+10, style loss: 1.2739e+07, content loss: 5.1581e+10, time: 3.9188s Iteration: 867 Total loss: 6.4304e+10, style loss: 1.2735e+07, content loss: 5.1568e+10, time: 3.9314s Iteration: 868 Total loss: 6.4288e+10, style loss: 1.2732e+07, content loss: 5.1555e+10, time: 3.9153s Iteration: 869 Total loss: 6.4272e+10, style loss: 1.2729e+07, content loss: 5.1543e+10, time: 3.9497s Iteration: 870 Total loss: 6.4256e+10, style loss: 1.2726e+07, content loss: 5.1531e+10, time: 3.9651s Iteration: 871 Total loss: 6.4241e+10, style loss: 1.2722e+07, content loss: 5.1519e+10, time: 3.9431s Iteration: 872 Total loss: 6.4225e+10, style loss: 1.2719e+07, content loss: 5.1506e+10, time: 3.9368s Iteration: 873 Total loss: 6.4210e+10, style loss: 1.2716e+07, content loss: 5.1494e+10, time: 3.9744s Iteration: 874 Total loss: 6.4194e+10, style loss: 1.2713e+07, content loss: 5.1481e+10, time: 3.9687s Iteration: 875 Total loss: 6.4179e+10, style loss: 1.2709e+07, content loss: 5.1470e+10, time: 3.9467s Iteration: 876 Total loss: 6.4164e+10, style loss: 1.2706e+07, content loss: 5.1458e+10, time: 3.9418s Iteration: 877 Total loss: 6.4148e+10, style loss: 1.2702e+07, content loss: 5.1446e+10, time: 3.9322s Iteration: 878 Total loss: 6.4133e+10, style loss: 1.2699e+07, content loss: 5.1434e+10, time: 3.9610s Iteration: 879 Total loss: 6.4118e+10, style loss: 1.2695e+07, content loss: 5.1423e+10, time: 3.9161s Iteration: 880 Total loss: 6.4103e+10, style loss: 1.2691e+07, content loss: 5.1412e+10, time: 3.9350s Iteration: 881 Total loss: 6.4088e+10, style loss: 1.2687e+07, content loss: 5.1401e+10, time: 3.9788s Iteration: 882 Total loss: 6.4072e+10, style loss: 1.2684e+07, content loss: 5.1388e+10, time: 3.9960s Iteration: 883 Total loss: 6.4057e+10, style loss: 1.2682e+07, content loss: 5.1375e+10, time: 3.9711s Iteration: 884 Total loss: 6.4042e+10, style loss: 1.2680e+07, content loss: 5.1363e+10, time: 3.9679s Iteration: 885 Total loss: 6.4028e+10, style loss: 1.2677e+07, content loss: 5.1351e+10, time: 3.9613s Iteration: 886 Total loss: 6.4013e+10, style loss: 1.2674e+07, content loss: 5.1339e+10, time: 3.8932s Iteration: 887 Total loss: 6.3998e+10, style loss: 1.2671e+07, content loss: 5.1327e+10, time: 3.9801s Iteration: 888 Total loss: 6.3983e+10, style loss: 1.2668e+07, content loss: 5.1315e+10, time: 4.0026s Iteration: 889 Total loss: 6.3969e+10, style loss: 1.2665e+07, content loss: 5.1304e+10, time: 3.9333s Iteration: 890 Total loss: 6.3954e+10, style loss: 1.2661e+07, content loss: 5.1293e+10, time: 3.9534s Iteration: 891 Total loss: 6.3939e+10, style loss: 1.2659e+07, content loss: 5.1281e+10, time: 3.9443s Iteration: 892 Total loss: 6.3925e+10, style loss: 1.2656e+07, content loss: 5.1269e+10, time: 3.9432s Iteration: 893 Total loss: 6.3911e+10, style loss: 1.2654e+07, content loss: 5.1257e+10, time: 4.0070s Iteration: 894 Total loss: 6.3896e+10, style loss: 1.2650e+07, content loss: 5.1246e+10, time: 3.9933s Iteration: 895 Total loss: 6.3882e+10, style loss: 1.2647e+07, content loss: 5.1235e+10, time: 3.9233s Iteration: 896 Total loss: 6.3868e+10, style loss: 1.2645e+07, content loss: 5.1223e+10, time: 3.9513s Iteration: 897 Total loss: 6.3854e+10, style loss: 1.2643e+07, content loss: 5.1211e+10, time: 3.9311s Iteration: 898 Total loss: 6.3839e+10, style loss: 1.2641e+07, content loss: 5.1198e+10, time: 3.9587s Iteration: 899 Total loss: 6.3825e+10, style loss: 1.2639e+07, content loss: 5.1187e+10, time: 3.9708s Iteration: 900 Total loss: 6.3811e+10, style loss: 1.2636e+07, content loss: 5.1175e+10, time: 3.9580s Iteration: 901 Total loss: 6.3797e+10, style loss: 1.2634e+07, content loss: 5.1163e+10, time: 4.0098s Iteration: 902 Total loss: 6.3783e+10, style loss: 1.2632e+07, content loss: 5.1151e+10, time: 3.9733s Iteration: 903 Total loss: 6.3769e+10, style loss: 1.2630e+07, content loss: 5.1139e+10, time: 4.0327s Iteration: 904 Total loss: 6.3755e+10, style loss: 1.2627e+07, content loss: 5.1128e+10, time: 4.0181s Iteration: 905 Total loss: 6.3741e+10, style loss: 1.2625e+07, content loss: 5.1116e+10, time: 3.9826s Iteration: 906 Total loss: 6.3727e+10, style loss: 1.2623e+07, content loss: 5.1104e+10, time: 3.9948s Iteration: 907 Total loss: 6.3713e+10, style loss: 1.2621e+07, content loss: 5.1092e+10, time: 3.9412s Iteration: 908 Total loss: 6.3699e+10, style loss: 1.2618e+07, content loss: 5.1081e+10, time: 3.9477s Iteration: 909 Total loss: 6.3686e+10, style loss: 1.2615e+07, content loss: 5.1071e+10, time: 3.9483s Iteration: 910 Total loss: 6.3672e+10, style loss: 1.2612e+07, content loss: 5.1060e+10, time: 3.9749s Iteration: 911 Total loss: 6.3658e+10, style loss: 1.2609e+07, content loss: 5.1050e+10, time: 4.0016s Iteration: 912 Total loss: 6.3645e+10, style loss: 1.2605e+07, content loss: 5.1039e+10, time: 3.9483s Iteration: 913 Total loss: 6.3631e+10, style loss: 1.2601e+07, content loss: 5.1030e+10, time: 3.9527s Iteration: 914 Total loss: 6.3617e+10, style loss: 1.2598e+07, content loss: 5.1020e+10, time: 3.9639s Iteration: 915 Total loss: 6.3604e+10, style loss: 1.2595e+07, content loss: 5.1009e+10, time: 3.9489s Iteration: 916 Total loss: 6.3590e+10, style loss: 1.2592e+07, content loss: 5.0998e+10, time: 3.9766s Iteration: 917 Total loss: 6.3577e+10, style loss: 1.2589e+07, content loss: 5.0988e+10, time: 3.9499s Iteration: 918 Total loss: 6.3564e+10, style loss: 1.2585e+07, content loss: 5.0979e+10, time: 3.9575s Iteration: 919 Total loss: 6.3550e+10, style loss: 1.2581e+07, content loss: 5.0969e+10, time: 3.9585s Iteration: 920 Total loss: 6.3537e+10, style loss: 1.2579e+07, content loss: 5.0958e+10, time: 3.9773s Iteration: 921 Total loss: 6.3524e+10, style loss: 1.2577e+07, content loss: 5.0947e+10, time: 3.9741s Iteration: 922 Total loss: 6.3511e+10, style loss: 1.2574e+07, content loss: 5.0937e+10, time: 3.9981s Iteration: 923 Total loss: 6.3498e+10, style loss: 1.2571e+07, content loss: 5.0926e+10, time: 4.0220s Iteration: 924 Total loss: 6.3484e+10, style loss: 1.2569e+07, content loss: 5.0916e+10, time: 4.0061s Iteration: 925 Total loss: 6.3471e+10, style loss: 1.2567e+07, content loss: 5.0904e+10, time: 3.9761s Iteration: 926 Total loss: 6.3458e+10, style loss: 1.2565e+07, content loss: 5.0894e+10, time: 3.9758s Iteration: 927 Total loss: 6.3445e+10, style loss: 1.2562e+07, content loss: 5.0884e+10, time: 3.9753s Iteration: 928 Total loss: 6.3432e+10, style loss: 1.2559e+07, content loss: 5.0873e+10, time: 4.0002s Iteration: 929 Total loss: 6.3419e+10, style loss: 1.2557e+07, content loss: 5.0862e+10, time: 3.9618s Iteration: 930 Total loss: 6.3406e+10, style loss: 1.2556e+07, content loss: 5.0850e+10, time: 3.9563s Iteration: 931 Total loss: 6.3393e+10, style loss: 1.2553e+07, content loss: 5.0840e+10, time: 3.9648s Iteration: 932 Total loss: 6.3380e+10, style loss: 1.2550e+07, content loss: 5.0831e+10, time: 3.9425s Iteration: 933 Total loss: 6.3367e+10, style loss: 1.2547e+07, content loss: 5.0821e+10, time: 3.9472s Iteration: 934 Total loss: 6.3355e+10, style loss: 1.2545e+07, content loss: 5.0810e+10, time: 3.9607s Iteration: 935 Total loss: 6.3342e+10, style loss: 1.2543e+07, content loss: 5.0799e+10, time: 4.0462s Iteration: 936 Total loss: 6.3329e+10, style loss: 1.2540e+07, content loss: 5.0789e+10, time: 4.0187s Iteration: 937 Total loss: 6.3317e+10, style loss: 1.2538e+07, content loss: 5.0778e+10, time: 4.0221s Iteration: 938 Total loss: 6.3304e+10, style loss: 1.2536e+07, content loss: 5.0768e+10, time: 3.9674s Iteration: 939 Total loss: 6.3291e+10, style loss: 1.2535e+07, content loss: 5.0757e+10, time: 3.9501s Iteration: 940 Total loss: 6.3279e+10, style loss: 1.2532e+07, content loss: 5.0747e+10, time: 3.9889s Iteration: 941 Total loss: 6.3266e+10, style loss: 1.2529e+07, content loss: 5.0738e+10, time: 3.9973s Iteration: 942 Total loss: 6.3254e+10, style loss: 1.2526e+07, content loss: 5.0728e+10, time: 4.0260s Iteration: 943 Total loss: 6.3242e+10, style loss: 1.2524e+07, content loss: 5.0718e+10, time: 4.0050s Iteration: 944 Total loss: 6.3229e+10, style loss: 1.2522e+07, content loss: 5.0708e+10, time: 3.9738s Iteration: 945 Total loss: 6.3217e+10, style loss: 1.2519e+07, content loss: 5.0698e+10, time: 3.9440s Iteration: 946 Total loss: 6.3205e+10, style loss: 1.2516e+07, content loss: 5.0689e+10, time: 3.9456s Iteration: 947 Total loss: 6.3193e+10, style loss: 1.2513e+07, content loss: 5.0679e+10, time: 3.9352s Iteration: 948 Total loss: 6.3181e+10, style loss: 1.2512e+07, content loss: 5.0669e+10, time: 3.9641s Iteration: 949 Total loss: 6.3168e+10, style loss: 1.2510e+07, content loss: 5.0659e+10, time: 3.9505s Iteration: 950 Total loss: 6.3156e+10, style loss: 1.2507e+07, content loss: 5.0650e+10, time: 3.9644s Iteration: 951 Total loss: 6.3145e+10, style loss: 1.2504e+07, content loss: 5.0640e+10, time: 3.9570s Iteration: 952 Total loss: 6.3133e+10, style loss: 1.2503e+07, content loss: 5.0630e+10, time: 3.9805s Iteration: 953 Total loss: 6.3121e+10, style loss: 1.2501e+07, content loss: 5.0620e+10, time: 3.9751s Iteration: 954 Total loss: 6.3109e+10, style loss: 1.2499e+07, content loss: 5.0610e+10, time: 3.9620s Iteration: 955 Total loss: 6.3097e+10, style loss: 1.2496e+07, content loss: 5.0601e+10, time: 3.9663s Iteration: 956 Total loss: 6.3085e+10, style loss: 1.2493e+07, content loss: 5.0592e+10, time: 3.9867s Iteration: 957 Total loss: 6.3073e+10, style loss: 1.2492e+07, content loss: 5.0582e+10, time: 4.0766s Iteration: 958 Total loss: 6.3062e+10, style loss: 1.2489e+07, content loss: 5.0573e+10, time: 4.0043s Iteration: 959 Total loss: 6.3050e+10, style loss: 1.2486e+07, content loss: 5.0564e+10, time: 3.9550s Iteration: 960 Total loss: 6.3038e+10, style loss: 1.2483e+07, content loss: 5.0556e+10, time: 3.9889s Iteration: 961 Total loss: 6.3027e+10, style loss: 1.2480e+07, content loss: 5.0547e+10, time: 4.0073s Iteration: 962 Total loss: 6.3015e+10, style loss: 1.2478e+07, content loss: 5.0537e+10, time: 3.9726s Iteration: 963 Total loss: 6.3004e+10, style loss: 1.2475e+07, content loss: 5.0529e+10, time: 4.0423s Iteration: 964 Total loss: 6.2992e+10, style loss: 1.2472e+07, content loss: 5.0520e+10, time: 4.0334s Iteration: 965 Total loss: 6.2981e+10, style loss: 1.2469e+07, content loss: 5.0512e+10, time: 3.9876s Iteration: 966 Total loss: 6.2969e+10, style loss: 1.2467e+07, content loss: 5.0502e+10, time: 3.9988s Iteration: 967 Total loss: 6.2958e+10, style loss: 1.2465e+07, content loss: 5.0493e+10, time: 4.0168s Iteration: 968 Total loss: 6.2946e+10, style loss: 1.2462e+07, content loss: 5.0485e+10, time: 4.0320s Iteration: 969 Total loss: 6.2935e+10, style loss: 1.2459e+07, content loss: 5.0476e+10, time: 4.0289s Iteration: 970 Total loss: 6.2924e+10, style loss: 1.2456e+07, content loss: 5.0468e+10, time: 4.0242s Iteration: 971 Total loss: 6.2912e+10, style loss: 1.2454e+07, content loss: 5.0459e+10, time: 4.0228s Iteration: 972 Total loss: 6.2901e+10, style loss: 1.2450e+07, content loss: 5.0450e+10, time: 4.0044s Iteration: 973 Total loss: 6.2889e+10, style loss: 1.2447e+07, content loss: 5.0443e+10, time: 4.0245s Iteration: 974 Total loss: 6.2878e+10, style loss: 1.2444e+07, content loss: 5.0434e+10, time: 3.9677s Iteration: 975 Total loss: 6.2867e+10, style loss: 1.2442e+07, content loss: 5.0425e+10, time: 3.9527s Iteration: 976 Total loss: 6.2856e+10, style loss: 1.2440e+07, content loss: 5.0416e+10, time: 3.9611s Iteration: 977 Total loss: 6.2844e+10, style loss: 1.2437e+07, content loss: 5.0408e+10, time: 3.9745s Iteration: 978 Total loss: 6.2833e+10, style loss: 1.2434e+07, content loss: 5.0399e+10, time: 3.9911s Iteration: 979 Total loss: 6.2822e+10, style loss: 1.2432e+07, content loss: 5.0390e+10, time: 3.9938s Iteration: 980 Total loss: 6.2811e+10, style loss: 1.2431e+07, content loss: 5.0380e+10, time: 4.0199s Iteration: 981 Total loss: 6.2800e+10, style loss: 1.2429e+07, content loss: 5.0371e+10, time: 4.0059s Iteration: 982 Total loss: 6.2789e+10, style loss: 1.2426e+07, content loss: 5.0362e+10, time: 3.9942s Iteration: 983 Total loss: 6.2778e+10, style loss: 1.2425e+07, content loss: 5.0353e+10, time: 3.9756s Iteration: 984 Total loss: 6.2767e+10, style loss: 1.2424e+07, content loss: 5.0343e+10, time: 3.9849s Iteration: 985 Total loss: 6.2756e+10, style loss: 1.2422e+07, content loss: 5.0334e+10, time: 3.9413s Iteration: 986 Total loss: 6.2745e+10, style loss: 1.2420e+07, content loss: 5.0326e+10, time: 3.9411s Iteration: 987 Total loss: 6.2735e+10, style loss: 1.2417e+07, content loss: 5.0317e+10, time: 3.9101s Iteration: 988 Total loss: 6.2724e+10, style loss: 1.2415e+07, content loss: 5.0309e+10, time: 3.9587s Iteration: 989 Total loss: 6.2713e+10, style loss: 1.2413e+07, content loss: 5.0300e+10, time: 3.9131s Iteration: 990 Total loss: 6.2702e+10, style loss: 1.2411e+07, content loss: 5.0292e+10, time: 3.9288s Iteration: 991 Total loss: 6.2692e+10, style loss: 1.2408e+07, content loss: 5.0284e+10, time: 3.9510s Iteration: 992 Total loss: 6.2681e+10, style loss: 1.2407e+07, content loss: 5.0275e+10, time: 3.9203s Iteration: 993 Total loss: 6.2670e+10, style loss: 1.2406e+07, content loss: 5.0265e+10, time: 3.9316s Iteration: 994 Total loss: 6.2660e+10, style loss: 1.2404e+07, content loss: 5.0256e+10, time: 3.9562s Iteration: 995 Total loss: 6.2649e+10, style loss: 1.2402e+07, content loss: 5.0247e+10, time: 3.9121s Iteration: 996 Total loss: 6.2639e+10, style loss: 1.2399e+07, content loss: 5.0239e+10, time: 3.9331s Iteration: 997 Total loss: 6.2628e+10, style loss: 1.2398e+07, content loss: 5.0231e+10, time: 3.9071s Iteration: 998 Total loss: 6.2618e+10, style loss: 1.2396e+07, content loss: 5.0222e+10, time: 3.9318s Iteration: 999 Total loss: 6.2607e+10, style loss: 1.2393e+07, content loss: 5.0214e+10, time: 3.8856s Total time: 3938.6745s
<Figure size 864x864 with 0 Axes>
Total time: 3942.4
In [23]:
for i in range(len(output)):
plt.figure(figsize=(10, 10))
plt.imshow(deprocess_img(output[i]))
plt.title(str((i+1)*100) + ' iteration Output Image')
plt.show()
In [24]:
plt.plot(best_loss)
plt.title('Total Loss')
plt.show()
In [25]:
plt.plot(style_score)
plt.title('Style Loss')
plt.show()
In [26]:
plt.plot(content_score)
plt.title('Content Loss')
plt.show()
By using different random seeds, content and style weight ratios, numbers of iterations and the content layers used in the model, we produced many style transferred pictures and we put many of them in our project final report.
Reference¶
TensorFlow. (2018, September 27). Neural Style Transfer: Creating Art with Deep Learning using tf.keras and eager execution. Retrieved from https://medium.com/tensorflow/neural-style-transfer-creating-art-with-deep-learning-using-tf-keras-and-eager-execution-7d541ac31398.